| V 1955 R-CODE SDK (for ERS-7M3) has been released. openrsupport@(Administrator) [2006/01/12-04:21] |
| Response : Edit : Delete |
|
R-CODE
SDK (Redist7_ver3.zip for ERS-7M3) has been released. Please use this version with ERS-7M3. https://openr.aibo.com/openr/eng/perm/rcode/download/download.php4 Thank you, AIBO SDE Support |
| V 1944 Problem getting connected under r-code... balam74 [2005/12/20-07:24] |
| | Response : Edit : Delete |
|
| I'm
lost. I kindly ask for help... I got an ERS-7M3 with Mind3. I've set up | the wireless connection (ad-hoc) with my laptop and everything goes perfectly | between the PC and Mind3. | I installed Rcode Ver2 in a new MS. I copied the WLANCONF.TXT file from the | working Mind3 into the new MS with Rcode. I also deleted the owner.txt file | and created the noauth.cfg file. I started up my Aibo, but I can't ping the | aibo's IP... what's wrong? The wifi blue led never turns on... | | Thank you very much. | |
| +- 1945 RE:Problem getting connected under r-code... openrsupport@(Administrator) [2005/12/20-10:09] |
| | Response : Edit : Delete |
|
| Thank
you very much for your interest in R-CODE. Unfortunately, AIBO ERS-7/M3 | WLAN isn't currently supported by R-CODE Ver2. We are working on adding WLAN | support for M3 now. An announcement will be made here when WLAN support is | available. We apologize for this delay. | | Thank you, | | AIBO SDE Support | |
| +- 1954 RE:Problem getting connected under r-code... openrsupport@(Administrator) [2006/01/12-04:19] |
| Response : Edit : Delete |
|
R-CODE
for the ERS-7M3 is now available. AIBO SDE Support |
| V 1953 ERS7 Mind 3 Color Calibration ? sacrenouille [2006/01/09-19:26] |
| Response : Email : Edit : Delete |
|
Hi I own a ERS7 with Mind 3. Using the Browser on AiboHack, I've noticed that Aibo has some predifined "colors" to recognise ball, aibone, faces and the like. I've been able to identify the file where this info is on the memory stick : OPEN-R/APP/DATA/P/TR.CT Since my Aibo is always confused when it comes to identify its Ball of Aibone on my brown floor, I had the idea to modify manually this file to better "calibrate" its predefined colors. But it seems that when I changed even just one byte, Aibo doesn't recognize anything at all. The colors I manually entered seems to be OK (I removed brown from the colors for the AIBOne recognition and left only red and pink) because I can see them clearly with the Aibo Browser. It might be that there is a checksum byte of some kind in the file. Anyway, I guess the calibration will be tricky. Did anyone already do that ? Is there a software that can help me with the task ? Is the whole thing possible at all ? Thanks ! |
| V 1952 Standalone charger ejt [2006/01/06-00:27] |
| Response : HomePage : Email : Edit : Delete |
|
It
appears the standalone charger for the ERS-7 has been discontinued. This means the only way to charge an ERS-7 battery is inside the robot, while it is sitting on the charger. For those of us doing research with these robots, that's not very convenient when you want to work for longer than the lifetime of a battery. Any suggestions? (or any standalone chargers for sale?) thanks, -ethan |
| V 1951 OPEN-R SDK file permissions ejt [2006/01/05-07:17] |
| Response : HomePage : Email : Edit : Delete |
|
I
just wanted to report that the file permissions set in the OPEN_R_SDK_1_1_5_r4.tar.gz distribution are a bit annoying, particularly in the case that OPEN_R_SDK is installed as 'root' on a linux or OS X machine. The problem is that files in OPEN_R/include/OPENR are set to only be readable by the owner (but not "group" or "other"), so that an installation by root is not usable unless permissions are changed (otherwise compilation fails with missing header files). A 'chmod -R a+r OPEN_R_SDK' fixes this problem. However, another problem that the comes up is that OPEN_R/bin/gzcp and OPEN_R/bin/stubgen2 do not have the 'executable' bit set for "group" or "other", so this also breaks builds unless permissions are changed again, this time with 'chmod -R a+x OPEN_R_SDK/OPEN_R/bin'. If there are any other problems, I have not come across them yet. thanks, -ethan |
| V 1947 Play WAV file with Aibo RFW randeep [2006/01/03-15:20] |
| | Response : Edit : Delete |
|
| How
to play a WAV file on the AIBO ERS7 using Aibo RFW when the WAV file is
on | the Aibo Memory Stick and when is is on the PC. Which APIs do I have to use? | |
| +- 1948 RE:Play WAV file with Aibo RFW AiboPet [2006/01/04-02:34] |
| | Response : Edit : Delete |
|
| >How
to play a WAV file on the AIBO ERS7 using Aibo RFW when the WAV file is
on | >the Aibo Memory Stick and when is is on the PC. Which APIs do I have to use? | | "AudioSendWAVFile" will upload (PC->AIBO) and play a single WAV file | [there is also AudioSendData/SendPlay/SendStop] | | If the WAV is a builtin one (on the AIBO), you should play it with the "MWCID" | command ID ("PlayContents") | | If you have a number of sounds you want to use, without uploading right before | playing, you can upload a custom ODA and MWC file ("FTPMMSend???File"). You | can also pre-install it on the Mind2/Mind3 stick to avoid the upload. This | gets a little tricky because you are creating your own MWCIDs (which must not | clash with the builtin ones) | |
| +- 1949 RE:Play WAV file with Aibo RFW randeep [2006/01/04-15:50] |
| | Response : Edit : Delete |
|
| Thanks
for your help. | Does it works with Mind1 too? Also, What format of the WAV file has to be used | I tested with 8kHz/ Mono and 8Khz/Stereo but does not seems to work :( | | >>How to play a WAV file on the AIBO ERS7 using Aibo RFW when the WAV file is | on | >>the Aibo Memory Stick and when is is on the PC. Which APIs do I have to use? | > | >"AudioSendWAVFile" will upload (PC->AIBO) and play a single WAV file | >[there is also AudioSendData/SendPlay/SendStop] | > | >If the WAV is a builtin one (on the AIBO), you should play it with the "MWCID" | >command ID ("PlayContents") | > | >If you have a number of sounds you want to use, without uploading right before | >playing, you can upload a custom ODA and MWC file ("FTPMMSend???File"). You | >can also pre-install it on the Mind2/Mind3 stick to avoid the upload. This | >gets a little tricky because you are creating your own MWCIDs (which must not | >clash with the builtin ones) | |
| +- 1950 RE:Play WAV file with Aibo RFW AiboPet [2006/01/05-01:48] |
| Response : Edit : Delete |
|
>Does
it works with Mind1 too? No. "RFW" was added with AiboEntertainmentPlayer (AEP-V1) in AiboMind2. (and continued and slightly improved in AEP-V2 with AiboMind3) >What format of the WAV file has to be used >I tested with 8kHz/ Mono and 8Khz/Stereo but does not seems to work :( It is very general - most regular formats should work (standard .WAV header/format) 8 or 16 bit, 8 or 16kHz, mono or stereo - all should work fine (and others) NOTE: you also need all the regular Sound playback initialization (much more than a single API call, including setting the format). I recommend copying the init code from one of the samples. "AudioInit", "AudioSetOpen" as well as "RequestNetService" for NETAUDIO_ID - and the order is important. [nothing is ever that simple ;-] |
| V 1946 Easy WLANCONF on urbiforge jcbaillie [2005/12/25-17:38] |
| Response : HomePage : Edit : Delete |
|
Hi, We have created a simple web front end to help people generate the WLANCONF.TXT file: http://www.urbiforge.com/wlanconf There are already programs to do this, but here since it's a web based application, it doesn't require to install anything. Hope it helps, Sincerely, JCB |
| V 1943 problem with compiling and linking - error page gohmshot [2005/12/19-02:55] |
| Response : Email : Edit : Delete |
|
I
have still none aibo. Property me the software downloaded. Compilen and
do not link go. the files mipsel linux gcc.exe and mipsel linux g .exe have 0 byte. and with compile and link of the examples comes the error message that something with draw-eat power monitor is not correct. As is this error to repair. this is an error on the openr.aibo.com page ! |
| V 1942 802.1 Authentication problem rapnroll [2005/12/18-11:52] |
| Response : Email : Edit : Delete |
|
Dear AIBO users: I have a difficulty to hook up my aibo to a commercial WiFi which requires 802.1 authentication with an ID/Password combination (not just a WEP key). I don't seem to find any relevant information of how to set up this kind of network. Is this impossible? Thanks |
| V 1936 GetSemInfo results aibo_user [2005/12/13-19:30] |
| | Response : Email : Edit : Delete |
|
| Hi to all, | | Has someone any info about what does the array "A" returned by GetSemInfo | contain? The only info from the RemoteTest sample is that | - A is an int[4] | - A[0] contains the serial ID of the last command executed | What about the other 3 items? | | A question in the question... | | How can I get the MWCID of the last action executed? | | Thanks in advance | | aibo_user | |
| +- 1938 RE:GetSemInfo results TodboT [2005/12/14-04:25] |
| | Response : Edit : Delete |
|
| >- A is an int[4] | >- A[0] contains the serial ID of the last command executed | >What about the other 3 items? | | I'm not sure what the other 3 values are. A lot of them are undocumented or | just not used. | | >How can I get the MWCID of the last action executed? | | I think you will have to keep track of the MWCID that you send. When you send | a motion command (like PlayMotion), the function will return the command ID. | Keep track of the command ID and associated MWCID. When you get SEM_RESULT, | match the command ID in A[0] to get the MWCID. | | TodboT | |
| +- 1941 RE:GetSemInfo results aibo_user [2005/12/15-23:46] |
| Response : Edit : Delete |
|
>>How
can I get the MWCID of the last action executed? > >I think you will have to keep track of the MWCID that you send. When you send >a motion command (like PlayMotion), the function will return the command ID. >Keep track of the command ID and associated MWCID. When you get SEM_RESULT, >match the command ID in A[0] to get the MWCID. Yes, this solution is OK to log user decided motion MWCIDs. But what about the others? A possibility can be using SEM_STATUS message, that (when AIBO's situation changes) returns its posture and status. Posture works, status returns weird results (too big numbers). Even using GetStatus() returns the same numbers, that don't match ST_STAT_ values... also in (Sony's!) RemoteTest sample... Any other idea? |
| V 1910 Run-time callstack code El Guapo [2005/11/08-16:17] |
| | Response : Email : Edit : Delete |
|
| I
have written a utility that will allow for callstacks to be displayed
via the | console at runtime. This can be used by any object at any time; all that is | required is an additional small build step than can be automated via the | makefile, and a single line of code to load symbols at startup. The included | readme file has exact instructions. Anyone who is interested in better | debugging, please try it out and let me know if it works! | | http://reactorsoftware.com/stuff/aibo_callstacks.0.9.tar.gz | | Also, is there any way add an exception handler for an object? Currently my | code will work via assertion or any other explicit call, but it would be great | to hook this up in the event of an exception. Sony? | | Thanks! | (The article that 2005/11/08-11:24 posted, was modified by Author) | |
| |- 1912 RE:Run-time callstack code ejt [2005/11/09-07:51] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | Very
exciting! I've always wanted something like this, but didn't know
enough of the low level stack | | and binary format to make it happen -- the only drawback for my own usage is that Tekkotsu is fairly | | large, and the memory footprint for all those functions is something to consider... (~1.5 MB -- not that | | bad really, but we're getting tight on memory on the 210 models, which we still use a few of in our lab, | | so I'm starting to get concerned about reducing memory usage.) | | | | Luckily, since most of the functions have common prefixes (e.g. 'char* std::basic_string<char, | | std::char_traits<char>, std::allocator<char> >'...), some kind of trie could do wonders for reducing the | | memory footprint. | | | | If you're interested in pursuing this, I have some ideas and code I could hand off, otherwise I might bug | | you off-list with a few implementation questions if I get around to pursuing it myself. (which might not | | happen -- probably will just end up adding it as an ERS-7 only feature for the near future.) | | | | Thanks again for making this available!!! | | -ethan | | |
| | +- 1913 RE:Run-time callstack code El Guapo [2005/11/10-03:28] |
| | Response : Email : Edit : Delete |
|
| Awesome,
glad this could be of use! There is a bit of extra overhead in the file | that could be reduced at the cost of a more complex build script; something that | I figured wouldn't be worth it since I wanted to keep everything as | straightforward as possible. I'd say maybe 5% wasted space, so if that would | make the difference in shoehorning it into a 210 than I may take another look. | | I'd like to persue more tools like this to make the robot a more fun platform to | code for. I may not have a lot of time in the near future, but I could certainly | take a look at your other items to see if I have some thoughts. | |
| +- 1930 RE: Sony info required! El Guapo [2005/12/04-01:50] |
| | Response : Email : Edit : Delete |
|
| Bump
for information from Sony about setting custom exception handlers. | Obviously there is a way.... | (The article that 2005/12/04-01:38 posted, was modified by Author) | |
| +- 1932 RE: Sony info required! openrsupport@(Administrator) [2005/12/08-11:10] |
| | Response : Edit : Delete |
|
| >Bump
for information from Sony about setting custom exception handlers. | >Obviously there is a way.... | | Thank you very much for sharing some of your work! | | Unfortunately, it is not possible to add your own user-made exception handler. | There is an exception monitor. EMON.CFG is the control file for the exception | monitor. The parameters are described in the ProgrammersGuide 5.2.5 EMON.CFG. | | Better debugging can be done if you use Remote Processing OPEN-R and gdb. | | Tutorials 6 and 7 may also help for debugging. | (https://openr.aibo.com/openr/eng/perm/university2.php4) | | 6. Debugging OPEN-R programs | We explain how to debug OPEN-R programs when CPU exceptions occurs. | Training course 6 (PDF) | | 7. Debugging by gdb using Remote Processing OPEN-R | We explain how to debug and trace OPEN-R programs by gdb using Remote Processing | OPEN-R. | Training course 7 (PDF) | | Thank you, | | AIBO SDE Support | |
| +- 1933 RE: Sony info required! El Guapo [2005/12/08-17:24] |
| | Response : Email : Edit : Delete |
|
| Thanks
for your response! | | >Unfortunately, it is not possible to add your own user-made exception handler. | >There is an exception monitor. EMON.CFG is the control file for the exception | >monitor. The parameters are described in the ProgrammersGuide 5.2.5 EMON.CFG. | | Is "not possible" is a mis-statement? It's software: anything is possible. I | guess you mean "not supported"? The system clearly allows a custom exception | handler object but you choose not to allow us to see how it works. Why not post | the source? | | >6. Debugging OPEN-R programs | >We explain how to debug OPEN-R programs when CPU exceptions occurs. | >Training course 6 (PDF) | | I'm aware of these documents, they helped me formulate the steps necessary to do | the same tracing at runtime. | | >7. Debugging by gdb using Remote Processing OPEN-R | >We explain how to debug and trace OPEN-R programs by gdb using Remote Processing | >OPEN-R. | >Training course 7 (PDF) | | My personal goal is to avoid the extra overhead of the remote processing...and | the endless list of steps you describe to get GDB running with a custom built | object look like a joke! I don't think its reasonable to expect any casual | programmer to go through all of that. | | I love the idea of an open robotics platform for hobbyists and applaud Sony for | allowing us access select tools and samples. At the same time I am shocked at | the user-hostile nature of the environment. It took me hours to get up and | running with even HelloWorld! When exceptions occur, tracing through assembly by | hand should not be the general purpose solution! | (The article that 2005/12/08-14:07 posted, was modified by Author) | |
| +- 1937 RE: Sony info required! openrsupport@(Administrator) [2005/12/14-04:00] |
| | Response : Edit : Delete |
|
| >Is
"not possible" is a mis-statement? It's software: anything is possible.
I | >guess you mean "not supported"? The system clearly allows a custom exception | >handler object but you choose not to allow us to see how it works. Why not post | >the source? | | Yes, you are correct. User exception handlers are not supported in the OPEN-R | SDK. Safety is a big concern of ours. When an exception does occur, we want to | make sure that the safety of the robot and operators is the highest priority. | For this reason, we do not allow user defined exception handlers or make | available the source code for this object. | | Thank you, | | AIBO SDE Support | |
| +- 1940 RE: Sony info required! ejt [2005/12/14-13:23] |
| Response : HomePage : Email : Edit : Delete |
|
>Yes,
you are correct. User exception handlers are not supported in the OPEN-R >SDK. Safety is a big concern of ours. When an exception does occur, we want to >make sure that the safety of the robot and operators is the highest priority. >For this reason, we do not allow user defined exception handlers or make >available the source code for this object. Are we talking about the same exceptions? I think El Guapo is interested in registering an abort() handler, for when a *C++ exception* is thrown and not caught. I'd also like to see the ability to catch a *CPU exception* (e.g. divide by zero) instead of just crashing. Note that these are different meanings and types of "exception". I have a feeling you are thinking of the *system exceptions* for "Joint Over Current" and such, which isn't what we're interested in. The system handling of that is just fine. Please make sure your information of the inability and concern of user "exception" handlers is regarding C++ exceptions (e.g. handling abort()), and perhaps CPU exceptions (really, is divide by zero worth crashing over? Who does that anymore?). We don't want the system exceptions -- I don't want liability for the hardware anymore than you do, but crashing for software bugs is about as bad as you can get -- anything we might do to recover is surely better than falling over dead! |
| V 1935 WiFi birman1 [2005/12/13-17:45] |
| | Response : Edit : Delete |
|
| Having
had great difficulties persuading my Aibo to join my local network, I | have discovered that the problem was down to the WEP key. My network key was a | random set of hex numbers, but repesented some non printable characters. This | works fine for both my computers and router, but not Aibo. | | Changing the WEP key such that its characters were within the alpha numeric | range of codes solved the problem and the Aibo is now happily on the network. | | I don't know if this restriction is commonly known, but I know many people | struggle with the networking so hope this info may be of use. | |
| +- 1939 RE:WiFi TodboT [2005/12/14-04:28] |
| Response : Edit : Delete |
|
If
you have an ERS-7, you can use WLAN Manager to set a hex WEP key. TodboT |
| V 1844 GCC 4.0.1 available (but need Sony to recompile SDK) ejt [2005/09/12-04:05] |
| | Response : HomePage : Email : Edit : Delete |
|
| I've
adapted a build-devtools script to configure gcc-4.0.1,
binutils-2.16.1, newlib-1.13.0 to build for | the AIBO. It will compile and build either the .bz2 or .gz compressed versions of each of these | packages, apply patches, configure and install for AIBO development. | | The resulting tool chain seems to work fine right up until linking, where ABI differences between 3.3 | and the later versions cause problems. This is partly due to some changes in the internals of library | implementation, but I expect even if we wrote adapters there would be runtime issues with the ABI | changes. (see 'gcc update II' thread where I was reporting progress) | | So, all that we need from Sony is for someone to run the provided script to install the updated tool | chain, and hit 'make' on the SDK sources for us. Then we'll be able to take advantage of a faster C++ | compiler, better optimizations, various fixes in the C++ parsing, precompiled header support, etc. | | build-devtools script: | http://ejt.no-ip.org/stuff/build-devtools-4.0.1.sh | patch for mkbin to handle change in gcc command line: | http://ejt.no-ip.org/stuff/mkbin.diff | | As an added bonus, here's another (unrelated) patch for TinyFTPD which lets you put comments in the | password file and fixes a variety of minor warnings: | http://ejt.no-ip.org/stuff/TinyFTPD.diff | | thanks, | -ethan | (The article that 2005/09/12-04:02 posted, was modified by Author) | |
| |- 1859 RE:GCC 4.0.1 available (but need Sony to recompile SDK) nlopes [2005/09/24-18:47] |
| | Response : Edit : Delete |
|
| Any news from Sony? | |
| +- 1934 RE:GCC 4.0.1 available (but need Sony to recompile SDK) nlopes [2005/12/08-22:36] |
| Response : Edit : Delete |
|
Yupyy!! I've completed a full build of the devtools with gcc 4.0.2. The trick is to use the gcc's option to use the g++ 3.3 ABI. My pc has linux/gentoo with gcc 3.4.4. the script (based on Ethan's): http://mega.ist.utl.pt/~ncpl/build-devtools- 4.0.2.sh Before running the script, you'll need to patch mkbin with http://ejt.no- ip.org/stuff/mkbin.diff Nuno |
| V 1928 How make two aibos reached synchronously while dancing? abc [2005/12/01-22:26] |
| | Response : Edit : Delete |
|
| How make two aibos reached synchronously while dancing? | Can using the R-CODE.R reach this purpose? | FOR ERS-7M2. | |
| +- 1931 RE:How make two aibos reached synchronously while dancing? AiboPet [2005/12/04-01:57] |
| Response : Edit : Delete |
|
>How
make two aibos reached synchronously while dancing? >Can using the R-CODE.R reach this purpose? >FOR ERS-7M2. I've given similar advice before. Please look at "Disco AIBO" http://www.aibohack.com/disco/#ERS7 If that is not exactly what you are looking for, please explain in detail. "What we've got here is... failure to communicate." |
| V 1929 How let SEARCH PINK BALL the HEAD_PAN angle grabs accurate abc [2005/12/01-22:41] |
| Response : Edit : Delete |
|
How
let AIBO at the SEARCH PINK BALL the HEAD_PAN angle value grabs at to reach the most accurat? Can using the R-CODE.R reach this purpose? FOR ERS7M2. (The article that 2005/12/01-22:40 posted, was modified by Author) |
| V 1919 aibo programs ers210 packman069 [2005/11/11-21:46] |
| | Response : Email : Edit : Delete |
|
| brand
new to the aibo scene, and, am blasted away by the size of it. can | anyone point me in the right direction to buy software for the ers 210. | otherwise, can someone show me a few of their own programs so i can get the | hang of making my own??? please please help me, very distressed.... | |
| +- 1921 RE:aibo programs ers210 AiboPet [2005/11/12-01:01] |
| | Response : HomePage : Edit : Delete |
|
| >brand
new to the aibo scene, and, am blasted away by the size of it. can | >anyone point me in the right direction to buy software for the ers 210. | First I suggest asking on a regular Aibo owner BBS, like aibo-life.org | http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi | [this is a developer BBS] | | Here's my suggestions for buying ERS-210 accessories: | http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi? | ubb=get_topic;f=1;t=002738 | | NOTE: most are discontinued and can be hard to find (eg: on eBay). The PMS | sticks are sold by Sony or Robotoys. | | ----- | >otherwise, can someone show me a few of their own programs so i can get the | >hang of making my own??? please please help me, very distressed.... | There are many options for ERS-210 programming. All need a "PMS" stick. | I suggest starting with YART, a simple drag-and-drop editor for RCODE programs. | http://www.aibohack.com/rcode/yart.htm | There are other tools available for more low level programming (using | the "OpenR SDK" or toolkits built on it) | |
| +- 1927 RE:aibo programs ers210 jcbaillie [2005/12/01-18:43] |
| Response : HomePage : Email : Edit : Delete |
|
You
should have a look at URBI: http://www.urbiforge.com And what people say about it: http://www.urbiforge.com/index.php?option=com_content&task=view&id=53&Itemid=96 Good luck! JC Baillie >>brand new to the aibo scene, and, am blasted away by the size of it. can >>anyone point me in the right direction to buy software for the ers 210. >First I suggest asking on a regular Aibo owner BBS, like aibo-life.org >http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi >[this is a developer BBS] > >Here's my suggestions for buying ERS-210 accessories: >http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi? >ubb=get_topic;f=1;t=002738 > >NOTE: most are discontinued and can be hard to find (eg: on eBay). The PMS >sticks are sold by Sony or Robotoys. > >----- >>otherwise, can someone show me a few of their own programs so i can get the >>hang of making my own??? please please help me, very distressed.... >There are many options for ERS-210 programming. All need a "PMS" stick. >I suggest starting with YART, a simple drag-and-drop editor for RCODE programs. >http://www.aibohack.com/rcode/yart.htm >There are other tools available for more low level programming (using >the "OpenR SDK" or toolkits built on it) |
| V 1926 Open-R gone off of memory stick deboy69 [2005/11/24-10:24] |
| Response : Email : Edit : Delete |
|
Some
how the memory stick got cleared and the Open-R files that were on it
are now gone. I just need to transfer that information back to the memory stick anyone know where i can get the software for it and step by step on how to do it thanks. ERS7M3 is the Aibo version |
| V 1924 how many kinds of onward actions in rcode? abc [2005/11/19-02:11] |
| | Response : Email : Edit : Delete |
|
| Have any other onward actions in addition to (WALK and | MOVE.MOVE.NORMAL and MOVE.MOVE.SLOW ) in rocde? | what is meaning of name of action in command-ers7-20040501.txt? | |
| +- 1925 RE:how many kinds of onward actions in rcode? AiboPet [2005/11/20-02:18] |
| Response : HomePage : Edit : Delete |
|
>Have
any other onward actions in addition to (WALK and >MOVE.MOVE.NORMAL and MOVE.MOVE.SLOW ) in rocde? Those are the official walking motions. Ones you can play with "PLAY ACTION ? name?" + optional args >what is meaning of name of action in command-ers7-20040501.txt? The documentation is minimal. Try them out and see for yourself. ------ There are many other undocumented actions that use the walk engine. You must play them using "PLAY MWCID ?number?" + optional args. Or you can add them to the "ACTION.AMS" file. Here is the complete list of MWCIDs that use the walk engine. Please do not ask what they do, they are undocumented. Try them. 600 = l_walk#walk_null = "CHGPOS.WALK.NORMAL" 601 = l_walk#walk_walk 602 = l_walk#walk_walk 603 = l_walk#walk_null = "STOP_WALK" 611 = l_walk#walk_walk 612 = l_walk#walk_walk 613 = l_walk#walk_walk 621 = l_walk#walk_walk = "MOVE.MOVE.SLOW" 622 = l_walk#walk_walk = official "WALK" = "MOVE.MOVE.NORMAL" 631 = l_walk#walk_walk 632 = l_walk#walk_walk 634 = l_walk#walk_walk 635 = l_walk#walk_walk 636 = l_walk#walk_walk 641 = l_walk#walk_walk 642 = l_walk#walk_walk 680 = l_walk#walk_kick 682 = l_walkt#walkt_touch 1426 = l_walk#walk_swing 1466 = l_walk#walk_swing 1467 = l_walk#walk_swing 1468 = l_walk#walk_swing 1469 = l_walk#walk_swing 1470 = l_walk#walk_swing 1471 = l_walk#walk_swing 1472 = l_walk#walk_swing 1473 = l_walk#walk_swing 1474 = l_walk#walk_swing 1475 = l_walk#walk_swing 1476 = l_walk#walk_swing 1477 = l_walk#walk_swing 1478 = l_walk#walk_swing 1479 = l_walk#walk_swing 1480 = l_walk#walk_swing 1481 = l_walk#walk_swing 1482 = l_walk#walk_walk 1485 = l_walk#walk_walk 1488 = l_walk#walk_walk 1491 = l_walk#walk_swing 1492 = l_walk#walk_movebody 1493 = l_walk#walk_walk 1494 = l_walk#walk_walk 1495 = l_walk#walk_walk 1496 = l_walk#walk_walk 1497 = l_walk#walk_walk 1498 = l_walk#walk_walk 1499 = l_walk#walk_walk 1500 = l_walkk#walkk_null 1501 = l_walkk#walkk_swing 1502 = l_walkk#walkk_swing 1503 = l_walkk#walkk_swing 1504 = l_walkk#walkk_swing 1505 = l_walkk#walkk_swing 1506 = l_walkk#walkk_swing 1507 = l_walkk#walkk_swing 1508 = l_walkk#walkk_swing 1509 = l_walkk#walkk_swing 1510 = l_walkk#walkk_swing 1511 = l_walkk#walkk_swing 1512 = l_walkk#walkk_swing 1513 = l_walkk#walkk_swing 1514 = l_walkk#walkk_swing 1515 = l_walkk#walkk_swing 1516 = l_walkk#walkk_swing 1517 = l_walkk#walkk_swing 1518 = l_walkk#walkk_swing 1519 = l_walk#walk_walk 1520 = l_walk#walk_walk 1521 = l_walkk#walkk_walk 1522 = l_walkk#walkk_walk 1523 = l_walk#walk_walk 1524 = l_walk#walk_walk 1525 = l_walkk#walkk_walk 1526 = l_walkk#walkk_walk 1527 = l_walk#walk_walk 1528 = l_walk#walk_walk 1529 = l_walkk#walkk_walk 1530 = l_walkk#walkk_walk 1531 = l_walkk#walkk_movebody 1532 = l_walkl#walkl_null 1533 = l_walkl#walkl_swing 1534 = l_walkl#walkl_walk 1535 = l_walkl#walkl_walk 1536 = l_walkl#walkl_walk 1537 = l_walkl#walkl_movebody 1538 = l_walkl#walkl_kick 1539 = l_walk#walk_walk 1540 = l_walk#walk_walk 1541 = l_walk#walk_walk 1542 = l_walk#walk_walk 1543 = l_walk#walk_walk 1544 = l_walk#walk_walk 1545 = l_walk#walk_walk 1546 = l_walk#walk_walk 1547 = l_walk#walk_walk 1548 = l_walk#walk_walk 1549 = l_walk#walk_walk 1550 = l_walk#walk_walk 1551 = l_walk#walk_walk 1552 = l_walk#walk_walk 1553 = l_walk#walk_walk 1554 = l_walk#walk_walk 1555 = l_walk#walk_walk 1556 = l_walkk#walkk_walk 1557 = l_walkk#walkk_walk 1558 = l_walkl#walkl_walk 1559 = l_walk#walk_walk 1560 = l_walk#walk_walk 1561 = l_walk#walk_walk 1562 = l_walkk#walkk_walk 1563 = l_walkk#walkk_walk 1564 = l_walkl#walkl_walk 1565 = l_walkk#walkk_null 1566 = l_walkl#walkl_null 1567 = l_walk#walk_movebody |
| V 1923 Problems with compiling "TakePicture" Sample kinek [2005/11/16-20:39] |
| Response : Email : Edit : Delete |
|
Hello, i tried to compile the "TakePicture" Sample with Microsoft Visual Studio .NET 2003 and get some problems. The current error ist: BaseClient fatal error LNK1181: cannot open input file 'c:\DIPLOM_STUFF\AIBO\AIBO.obj' There is no file AIBO.obj on my PC. Maybee Visual Studio cant$B%((Bt find some DLLs? I$B%((Bm a newbie in programming with Visual Studio. can anybody help? thanks in advance kinek |
| V 1914 How to use the (CHGPOS.WALK.NORMAL) in rcode-ers 7th- E.txt? abc [2005/11/10-21:22] |
| | Response : Email : Edit : Delete |
|
| How
to use the (CHGPOS.WALK.NORMAL)Change to walking posture in rcode-ers
7th- | E.txt? | for ers-7. | thanks for help. | (The article that 2005/11/10-21:18 posted, was modified by Author) | |
| +- 1917 RE:How to use the (CHGPOS.WALK.NORMAL) in rcode-ers 7th- E.txt? AiboPet [2005/11/11-01:39] |
| | Response : HomePage : Edit : Delete |
|
| >How
to use the (CHGPOS.WALK.NORMAL)Change to walking posture in rcode-ers
7th- | >E.txt? | >for ers-7. | | Like most of the basic motions, | PLAY ACTION action_name | so in this case: | PLAY ACTION CHGPOS.WALK.NORMAL | |
| +- 1918 RE:How to use the (CHGPOS.WALK.NORMAL) in rcode-ers 7th- E.txt? abc [2005/11/11-15:32] |
| | Response : Email : Edit : Delete |
|
| >i
try to use you way just like (PLAY ACTION:CHGPOS.WALK.NORMAL) | but aibo didn't move .what qusetion for it? | what is (CHGPOS.WALK.NORMAL) main function? | (The article that 2005/11/11-15:28 posted, was modified by Author) | |
| +- 1920 RE:How to use the (CHGPOS.WALK.NORMAL) in rcode-ers 7th- E.txt? AiboPet [2005/11/12-00:54] |
| | Response : HomePage : Edit : Delete |
|
| >what
is (CHGPOS.WALK.NORMAL) main function? | | "CHGPOS" means change posture | | "CHGPOS.WALK.NORMAL" is for moving AIBO into a position to start walking. It | does not walk. Use the regular WALK action to make him walk. | | You don't need to use "CHGPOS.WALK.NORMAL". AIBO will move to that posture | when it needs to before walking. | |
| +- 1922 RE:How to use the (CHGPOS.WALK.NORMAL) in rcode-ers 7th- E.txt? abc [2005/11/15-18:19] |
| Response : Email : Edit : Delete |
|
Have
any other onward actions in addition to (WALK and MOVE.MOVE.NORMAL and MOVE.MOVE.SLOW ) in rocde? what is meaning of name of action in command-ers7-20040501.txt? (The article that 2005/11/12-03:30 posted, was modified by Author) |
| V 1915 TROUBLE INSTALLING R-CODE ON ERS-7 AIBO ebrozows [2005/11/10-23:55] |
| | Response : Email : Edit : Delete |
|
| After
applying the zip file to the root of my MIND2 memory stick, the AIBO
robot | does not initialize properly. (It justs turns on and then back off.) I have | tried ever possible approach to the situation and I am lost. My main goal is to | be able to access the telnet feature on port 21002 after installing the RCODE | zip file. (Redist7_ver2.zip to be exact.) | | Also, I am able to communicate wirelessly through the AIBO entertainment player. | | How am supposed to open up port 21002? | | Why wont the RCODE file work? | (The article that 2005/11/10-23:54 posted, was modified by Author) | |
| +- 1916 RE:TROUBLE INSTALLING R-CODE ON ERS-7 AIBO AiboPet [2005/11/11-01:36] |
| Response : HomePage : Edit : Delete |
|
RCODE
will not run on the AiboMind memory stick! You need a "PMS" stick (8 or 16MB) Read more about it here: http://www.aibohack.com/pms.htm ---- The 32MB AiboMind stick is only good for running AiboMind1, AiboMind2 or a very trimmed down version of AiboMind3. |
| V 1911 AIBO Remote Framework Tutorial kinek [2005/11/08-22:57] |
| Response : Email : Edit : Delete |
|
Hello All, I am looking for a tutorial about programming with "AIBO Remote Framework". I have found a lot of stuff about Open-R on this site, but only a litte about pragramming with "AIBO Remote Framework". There is the API and the sample programms, but i can not find some conceptional tutorials. Is there somthing like that on the web? thank in advance for helping tomasz gingold berlin (The article that 2005/11/08-22:37 posted, was modified by Author) |
| V 1906 Error measures obtained when using GetSensorValue in OPENR outsider [2005/11/03-22:19] |
| | Response : HomePage : Edit : Delete |
|
| Hello everybody, | I'm trying to retrieve the value of the long distance sensor in Aibo ERS7, using | the OPENR primitive GetSensorValue. | I perform a measure of that sensor using that primitive every 96 ms, but I | receive a confusing answer. The first ten or eleven measures are correct, but | then I receive ten or eleven measures with 1500000, then ten or eleven measures | correct, and then ten or eleven with 1500000, and so on. | I do not understand why all the measures are not correct. | Ihave changed the step of measuring to 32 ms and then I obtain 2 measures | correct and one measure 1500000, and so on. | Anybody has any idea of what is happening?. Why I do not obtain a correct | measure at such a high time-step?. | | Thank you | | Ricardo | |
| +- 1908 RE:Error measures obtained when using GetSensorValue in OPENR TodboT [2005/11/04-03:07] |
| | Response : Edit : Delete |
|
| It
looks like you're using an ers-7. There are two distance sensors in the | nose of the robot, so the distance reported from each sensor is interleaved. | Look at ModelInformation_7_E.pdf. I haven't used GetSensorValue with the PSD | sensors, so I'm not sure of the behavior. I've only read the values from an | OSensorFrameVectorData. | | TodboT | |
| +- 1909 More data but still not clear how it works outsider [2005/11/04-19:32] |
| Response : HomePage : Edit : Delete |
|
>It
looks like you're using an ers-7. There are two distance sensors in the >nose of the robot, so the distance reported from each sensor is interleaved. >Look at ModelInformation_7_E.pdf. I haven't used GetSensorValue with the PSD >sensors, so I'm not sure of the behavior. I've only read the values from an >OSensorFrameVectorData. > >TodboT Yes I'm using an ERS-7 model. As you say, there are two distance sensors, and their measurements are interleaved at a 50 ms rate. I understand how it works with the OSensorFrameVectorData, but I would like to perform a 'I don't care' measure of the distance. By 'I don't care' measure I mean just call a function and receive the measured value without handling all the structure data of OSensorFrameVectorData. By looking at the Reference guide, the OPENR::GetSensorValue function promises that (a 'I don't care' measure). But it doesn't work that way. So I'm trying to figure out how it works. Anybody can explain me that? Any comment will be welcome. Thanks Ricardo |
| V 1891 How to stop custom motions ferrante [2005/10/27-22:55] |
| | Response : HomePage : Email : Edit : Delete |
|
| I've
a problem with Remote Framework's FTP: once a custom movement has been | correctly loaded with FTP and launched with PlayContents, how can I stop it? | The infinity flag in MWC.CFG is correctly 0x01 (stop allowed); the wait flag | in PlayContents is correctly "false". Why, when I try to stop motion with | MotionStop(), PlayContents(2993, false) or StopRobotFunction() the motion | doesn't stop? | Where's the mistake? | |
| +- 1907 RE:How to stop custom motions aibo_user [2005/11/04-00:42] |
| Response : Edit : Delete |
|
Last update: neither aibo->SendCmd(APPCMD_PLAY_STOP, false) seems to work... Please, help us! |
| V 1902 OPEN-R-SDK build problem with Mac OS 10.4 Tiger raph [2005/11/01-23:38] |
| | Response : Edit : Delete |
|
| Hello, | | I am trying to build the OPEN-R-SDK on a Mac with Mac OS X Tiger (10.4.3). It always fails after a while | with an "internal compiler error", regardless which version of XTools (and gcc) I use. | | Here is what I exactly did: | | I followed the exact instructions from the installation guide from the documentation package. I put all | the needed tar.gz in a directory where the shell script | | build-devtools-3.3.2-macosx-r1.sh | | is. I launch it as a user (but I also tried as root) with | | ./build-devtools-3.3.2-macosx-r1.sh | | It always fails at some point of the compilation of the patched gcc-3.3.2. I tried with gcc-4.2 from | XTools 2.0, with gcc-2.95.2, gcc-3.3 and gcc-3.1 from XTools 1.1. The error message is not always the | same, but it is always an "internal compiler error" (except for gcc 2.95.2 where it was an syntax error). | | I would be really thankfull for any kind of help | | Best regards, | | raph | |
| +- 1905 RE:OPEN-R-SDK build problem with Mac OS 10.4 Tiger raph [2005/11/03-21:53] |
| Response : Edit : Delete |
|
Hello, >It always fails at some point of the compilation of the patched gcc-3.3.2. I tried with gcc-4.2 from >XTools 2.0, with gcc-2.95.2, gcc-3.3 and gcc-3.1 from XTools 1.1. The error message is not always the I finaly got it working with gcc-3.1 Best regards raph |
| V 1904 Error measures obtained when using GetSensorValue in OPENR outsider [2005/11/03-21:53] |
| Response : HomePage : Edit : Delete |
|
Hello everybody, I'm trying to retrieve the value of the long distance sensor in Aibo ERS7, using the OPENR primitive GetSensorValue. I perform a measure of that sensor using that primitive every 96 ms, but I receive a confusing answer. The first ten or eleven measures are correct, but then I receive ten or eleven measures with 1500000, then ten or eleven measures correct, and then ten or eleven with 1500000, and so on. I do not understand why all the measures are not correct. Ihave changed the step of measuring to 32 ms and then I obtain 2 measures correct and one measure 1500000, and so on. Anybody has any idea of what is happening?. Why I do not obtain a correct measure at such a high time-step?. Thank you Ricardo |
| V 1894 Garbage characters in disassembly? El Guapo [2005/10/31-06:45] |
| | Response : Edit : Delete |
|
| When
disassembling my object, I see a bunch of random characters mixed in
with | my function names. What's going on here? | | This is what I see instead of | test2() -> _Z5test2v | MainObj::DoInit(OSystemEvent) -> _ZN8CMainObj6DoInitERK12OSystemEvent | | I am seeing the same thing when I try to generate a map file as well. What build | step is converting the function names into this format? Is there a way for me to | undo it? (A spec that describes how special characters should be replaced?) | | EDIT: It looks like this is how the names appear in the .o files, meaning | mipsel-linux-g++ is doing it. Still, I need to know how to get the original names. | (The article that 2005/10/31-04:11 posted, was modified by Author) | |
| |- 1895 RE:Garbage characters in disassembly? (mostly fixed) El Guapo [2005/10/31-17:36] |
| | Response : Edit : Delete |
|
| FYI
the executable c++filt will do this, although it is not perfect... | | >When disassembling my object, I see a bunch of random characters mixed in with | >my function names. What's going on here? | > | >This is what I see instead of | >test2() -> _Z5test2v | >MainObj::DoInit(OSystemEvent) -> _ZN8CMainObj6DoInitERK12OSystemEvent | > | >I am seeing the same thing when I try to generate a map file as well. What build | >step is converting the function names into this format? Is there a way for me to | >undo it? (A spec that describes how special characters should be replaced?) | > | >EDIT: It looks like this is how the names appear in the .o files, meaning | >mipsel-linux-g++ is doing it. Still, I need to know how to get the original names. | > | (The article that 2005/10/31-16:34 posted, was modified by Author) | |
| |- 1897 RE:Garbage characters in disassembly? AiboPet [2005/11/01-01:13] |
| | | Response : HomePage : Edit : Delete |
|
| | >When
disassembling my object, I see a bunch of random characters mixed in
with | | >my function names. What's going on here? | | | | That's the way most C++ compilers work - the mangled name used by the linker | | includes the argument types, return types etc | | That's so you can do: | | | | int foo(int one_arg); | | int foo(int one_arg, int two_arg); | | int foo(char* string_arg); | | | | And all are named "foo", but have 3 different linker names. If you use the | | Visual C++ debugger, it will display the original names for you. | | |
| | +- 1903 RE:Garbage characters in disassembly? El Guapo [2005/11/02-04:22] |
| | Response : Edit : Delete |
|
| Thanks
AiboPet and wolftag. I didn't realize that the names would be output in | that format in the debug ELF. I'm using c++filt to restore them now. This is for | a small project that will hopefully be of use to the community, so I'll keep you | posted | | >>When disassembling my object, I see a bunch of random characters mixed in with | >>my function names. What's going on here? | > | >That's the way most C++ compilers work - the mangled name used by the linker | >includes the argument types, return types etc | >That's so you can do: | > | >int foo(int one_arg); | >int foo(int one_arg, int two_arg); | >int foo(char* string_arg); | > | >And all are named "foo", but have 3 different linker names. If you use the | >Visual C++ debugger, it will display the original names for you. | |
| +- 1900 RE:Garbage characters in disassembly? wolftag [2005/11/01-18:00] |
| | Response : Edit : Delete |
|
| >When
disassembling my object, I see a bunch of random characters mixed in
with | >my function names. What's going on here? | > | >This is what I see instead of | >test2() -> _Z5test2v | >MainObj::DoInit(OSystemEvent) -> _ZN8CMainObj6DoInitERK12OSystemEvent | > | >I am seeing the same thing when I try to generate a map file as well. What | build | >step is converting the function names into this format? Is there a way for me | to | >undo it? (A spec that describes how special characters should be replaced?) | > | >EDIT: It looks like this is how the names appear in the .o files, meaning | >mipsel-linux-g++ is doing it. Still, I need to know how to get the original | names. | >(The article that 2005/10/31-04:11 posted, was modified by Author) | | As someone has already explained, that's C++ name mangling. If you have a few | symbols that you do not want mangled, wrap the definitions like this for C | style linkage: | | extern "C" { | | //definitions | | } | |
| +- 1901 RE:Garbage characters in disassembly? wolftag [2005/11/01-18:20] |
| Response : Edit : Delete |
|
>>When
disassembling my object, I see a bunch of random characters mixed in with >>my function names. What's going on here? >> >>This is what I see instead of >>test2() -> _Z5test2v >>MainObj::DoInit(OSystemEvent) -> _ZN8CMainObj6DoInitERK12OSystemEvent >> >>I am seeing the same thing when I try to generate a map file as well. What >build >>step is converting the function names into this format? Is there a way for me >to >>undo it? (A spec that describes how special characters should be replaced?) >> >>EDIT: It looks like this is how the names appear in the .o files, meaning >>mipsel-linux-g++ is doing it. Still, I need to know how to get the original >names. >>(The article that 2005/10/31-04:11 posted, was modified by Author) > >As someone has already explained, that's C++ name mangling. If you have a few >symbols that you do not want mangled, wrap the definitions like this for C >style linkage: > >extern "C" { > >//definitions > >} Sorry, I should have said declaration, not definition. So: extern "C" { int myFunc(int x, int y); } |
| V 1893 Could you please give me the ers-7 walk (mtn,kfm,pose)file abc [2005/10/31-23:56] |
| | Response : Email : Edit : Delete |
|
| Could
you please give me the kfm file and the pose file and mtn file that the | ers-7 walk? | i am student in taiwan. I just for study about aibo technology and knowledge. | hope you can help me. | thanks for help. | (The article that 2005/10/31-23:54 posted, was modified by Author) | |
| +- 1899 RE:Could you please give me the ers-7 walk (mtn,kfm,pose)file AiboPet [2005/11/01-01:22] |
| Response : HomePage : Edit : Delete |
|
>Could
you please give me the kfm file and the pose file and mtn file that the >ers-7 walk? You've asked this question a few times already. It is difficult to provide more information with such cryptic requests. Please describe what you are trying to do, what tools, and what software you are using right now. Then it may be possible to be more helpful. The regular ERS-7 walks do not use .MTN files There is a simple sample walk motion in the OpenR SDK MoNet sample - meant for the OpenR SDK (not RCODE) |
| V 1896 Cookbook for Aibo Lan on Mac OSX 10.3.9 required montgrimpulo [2005/11/01-00:30] |
| | Response : Email : Edit : Delete |
|
| Sorry, this old story again ... | Can somebody send me a step by step installation instruction, how to configure an Aibo network in | system preferences (network) Mac OSX 10.3.9, corresponding to my wlanconf.txt: | | HOSTNAME=AIBO | ETHER_IP=10.0.1.100 | ETHER_NETMASK=255.255.255.0 | IP_GATEWAY=10.0.1.1 | ESSID=AIBONET | WEPENABLE=0 | WEPKEY= | APMODE=2 | CHANNEL=11 | | My Mac is connected via an Ethernet-cable to the apple airport. Any additional aiport configuration | needed ? Which telnet command may I use to connect to Aibo ? What is the sequence to set up a | connection? | | Best, | | montgrimpulo | |
| +- 1898 RE:Cookbook for Aibo Lan on Mac OSX 10.3.9 required AiboPet [2005/11/01-01:16] |
| Response : HomePage : Edit : Delete |
|
Here's
one guide (a little out of date, but shouldn't have changed much) http://www.droidlogic.com/aibo/open-r/OSX.html |
| V 1886 TinyFTPD connection flakiness El Guapo [2005/10/26-04:43] |
| | Response : Email : Edit : Delete |
|
| Hi
all! I am beginning my experimentation with the Open-R SDK and have run
into | a small problem, and was hoping someone who has experienced this could shed some | light on it. | | My first step is to automate the build process and allow it to be done remotely. | To do this, I am using the included TinyFTPD module. I have some scripts which | automate the FTP session, generating a series of steps which look something like: | | quote mkd CONF | cd CONF | type binary | put sample/common/tinyftpd/ms/OPEN-R/MW/CONF/OBJECT.CFG OBJECT.CFG | | etc etc, which are then piped into the unix ftp client. I have also tried using | the included MSTREEPUT script with approximately the same success rate. I have | tested these session scripts with my own FTP server on the internet with 100% | success rate from several different machines (all running WindowsXP with | Cygwin). The issue comes when I try connecting to the server running on AIBO. | | When I build TinyFTPD in final mode, about 40% of the time a file transfer will | hang right before beginning. It behaves as if some packets were dropped and it | wasn't using TCP; the transfer process merely sits idle until I kill it. | | When I build TinyFTPD in debug mode (after fixing some compile errors, shame) | and watch the debug output via telnet, suddenly all my transfers work great, if | a bit slow | | It seems there is a timing issue related to send/receive buffers, but I'm not | really sure if the problem is with the TCP/IP stack or the TinyFTPD | implementation, or where to start debugging this. Any insight would be | appreciated! Thanks. | | [edit] I'm running SDK version 1.1.5-r3 on ERS-7 | (The article that 2005/10/26-02:53 posted, was modified by Author) | |
| +- 1889 RE:TinyFTPD connection flakiness El Guapo [2005/10/28-01:20] |
| Response : Edit : Delete |
|
Here are the issues I've found. 1) The FtpDTP FILE* member (fp) is opened when a transfer request is initiated, but not closed unless the transfer successfully completes. This will make subsequent attempts to overwrite the file fail (and then you have to take out the memory stick...) Solution: Only open file handle on the first successful incoming data packet. Close it (if open) in FtpDTP::Close() in addition to the current locations. 2) If a connection is dropped at the beginning of FtpDTP::Store(), it may go undetected and result in extended periods of inactivity. Made worse by (1) in which a file would be left locked. Solution: Propagate back error codes! Several functions in this chain do not bother to check errors Having made these changes locally I have found a huge improvement in the reliability of the program. I would strongly recommend making these changes if you use it a lot. If anyone has similar points that they've improved, I'm all ears! My fixes don't address the low level issue of connection problems, but they at least make the program's response bearable. (The article that 2005/10/27-13:45 posted, was modified by Author) |
| V 1892 How to undo ImageSetWindowHandle aibo_user [2005/10/28-00:40] |
| Response : Edit : Delete |
|
Perhaps a stupid, but complicated question... To receive image data, I start ImageTP service, receiving WM_IMAGETP_UPDATE messages; at this point, I can grab image data with ImageGetData. To stop receiving image data, I stop ImageTP service. If I start ImageTP service and then use ImageSetWindowHandle, I can choose a window where data must be shown. If I stop ImageTP service, image display in the window is automatically stopped. And now, the problem: How can I undo ImageSetWindowHandle effects? In other words, how can I get a behaviour similar to the image on/off button of Aibo Player *without* stop ImageTP service? |
| V 1890 How writes r-code let aibo increment ers-7 walked speed? abc [2005/10/27-20:31] |
| Response : Email : Edit : Delete |
|
How writes r-code let aibo increment ers-7 walked speed? |
| V 1887 Is There the parameter that can let to walk faster IN R-CODE abc [2005/10/26-21:27] |
| | Response : Email : Edit : Delete |
|
| Is
There the parameter that can let to walk faster IN R-CODE.R? | |
| +- 1888 RE:Is There the parameter that can let to walk faster IN R-CODE AiboPet [2005/10/27-01:05] |
| Response : HomePage : Edit : Delete |
|
Please
remind me what model AIBO are you using ? >Is There the parameter that can let to walk faster IN R-CODE.R? If using RCodePlus (ERS-210,220,31x), there are extensions to the basic "WALK" command in the format "WALK.STYLEn" 2 is normal, 1 is slow, 3 is fast. eg: PLAY ACTION WALK.STYLE3 0 1000 See the AiboRemote program for the other walking styles. ---- A similar technique may be possible with ERS-7 RCODE (you need some tricky editing of the .MWC file). I don't think there are built-in command names for them. |
| V 1881 How can let several mtn files become an oda file? abc [2005/10/24-00:42] |
| | Response : Edit : Delete |
|
| Previous
I use RTOOL this software and only can let a mtn file become an oda | file . | How can let several mtn files become an oda file? | |
| +- 1885 RE:How can let several mtn files become an oda file? AiboPet [2005/10/26-01:42] |
| Response : HomePage : Edit : Delete |
|
>Previous
I use RTOOL this software and only can let a mtn file become an oda >file . >How can let several mtn files become an oda file? In the RTOOL "ODA" page, check the "MOTION Data" option and point it to the folder containing several MTN files (default ".\ODA\MOTION"). Press "Make ODA" and it will combine them and save them as MOTION.ODA (in the specified Output Path) |
| V 1882 [RFW] Sockets in Samples - what for? Shay [2005/10/25-00:36] |
| | Response : Edit : Delete |
|
| There
is a server socket in TakePicture sample. But I can't understand what is | it for? It seems it doesn't send any interesting information except "OK" | or "NG". | |
| +- 1884 RE:[RFW] Sockets in Samples - what for? AiboPet [2005/10/26-01:37] |
| Response : HomePage : Edit : Delete |
|
>There
is a server socket in TakePicture sample. But I can't understand what is >it for? It seems it doesn't send any interesting information except "OK" >or "NG". These are meant to be the AIBO controller for PHP webpages. The "NetControl" sample provides the shell of a PHP/server socket app - no real actions. The "TakePicture" sample has more commands (see RequestSocket.cpp), eg: HELLO, FORWARD, PHOTO etc See the two PHP scripts to see how these are called (customize for your web server). If not using PHP, you can roll your own interface. |
| V 1883 OPEN-R SDK 1.1.5-r4 (for ERS-7M3) has been released. openrsupport@(Administrator) [2005/10/25-16:07] |
| Response : Edit : Delete |
|
OPEN-R SDK 1.1.5-r4 (for ERS-7M3) has been released. Please use this version with ERS-7M3. There is no change to the specification and functions. https://openr.aibo.com/openr/eng/perm/openrsdk/download/download.php4 Thank you, AIBO SDE Support |
| V 1851 looking for commented sample code (OPEN-R ERS7) chicoisl [2005/09/22-04:41] |
| | Response : Edit : Delete |
|
| Hi there. | | I've been going through the Sony sample AIBO code, and am finding it hard to decipher due to the | complete lack of comments (was it auto-generated?). It also doesn't help that the code seems to | reference variables and functions that I can't find declarations for in any file/header/library. | | Anyway, I was wondering if anyone knows of any nicely commented, simple sample programs (beyond | hello world) that would be better suited to use as a starting point. | | Thanks for any help. | (The article that 2005/09/22-04:38 posted, was modified by Author) | |
| +- 1852 RE:looking for commented sample code jcbaillie [2005/09/15-16:02] |
| | Response : HomePage : Edit : Delete |
|
| You
should try the OPENR tutorial from ENSTA: | | http://uei.ensta.fr/baillie/eng/openr_tutorial.html | | Or even better: use URBI instead of OPENR: http://www.urbiforge.com | | Regards, | jcbaillie | | >Hi there. | > | >I've been going through the Sony sample AIBO code, and am finding it hard to | decipher due to the | >complete lack of comments (was it auto-generated?). It also doesn't help that | the code seems to | >reference variables and functions that I can't find declarations for in any | file/header/library. | > | >Anyway, I was wondering if anyone knows of any nicely commented, simple sample | programs (beyond | >hello world) that would be better suited to use as a starting point. | > | >Thanks for any help. | >(The article that 2005/09/15-06:34 posted, was modified by Author) | |
| +- 1853 RE:looking for commented sample code chicoisl [2005/09/20-04:03] |
| | Response : Edit : Delete |
|
| Thanks,
I'll look into that. | | Also, if anyone else has some nicely commented sample code, I'd still like to see some other examples. | | -Alex | | ---- | >You should try the OPENR tutorial from ENSTA: | > | >http://uei.ensta.fr/baillie/eng/openr_tutorial.html | > | >Or even better: use URBI instead of OPENR: http://www.urbiforge.com | > | >Regards, | >jcbaillie | (The article that 2005/09/20-04:02 posted, was modified by Author) | |
| +- 1854 RE:looking for commented sample code ejt [2005/09/20-06:23] |
| | Response : HomePage : Edit : Delete |
|
| This
is probably overkill for a request for "sample" code, but you can peruse | the Aperios-specific portion of Tekkotsu here: | http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/aperios/ | The "Capitalized" names are Aperios OObjects -- the others are 3rd party | libraries I've ported to Aperios. | | Of course, the whole point of Tekkotsu is the complexity of the Aperios | interaction is handled for you, so from a Tekkotsu perspective users aren't | expected to be needing to see that code. However, if you're looking to see some | more complicated/intricate/fancy usage, then it might help. | | (more on Tekkotsu here if you haven't already: http://www.tekkotsu.org ) | | -ethan | (The article that 2005/09/20-06:22 posted, was modified by Author) | |
| +- 1857 RE:looking for commented sample code (OPEN-R ERS7) chicoisl [2005/09/22-13:23] |
| | Response : Edit : Delete |
|
| I
guess I should have mentioned that I am trying to get my university
started down the RoboCup path. | We are a small team, and at this point we only have the manpower to work on one aspect of the code | while using code from elsewhere to control other functions. For that reason, compatibility with other | code is important. Would we be limiting ourselves in that respect with Tekkotsu (or URBI)? | | Also, I noticed that jcbaille's tutorial is for an ERS200 AIBO. It's quite helpful, but the code no longer | matches up with the ERS7 code. Is there anything else out there? | | Thanks for your help so far, | -Alex | | > | > | >This is probably overkill for a request for "sample" code, but you can peruse | >the Aperios-specific portion of Tekkotsu here: | > http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/aperios/ | >The "Capitalized" names are Aperios OObjects -- the others are 3rd party | >libraries I've ported to Aperios. | > | >Of course, the whole point of Tekkotsu is the complexity of the Aperios | >interaction is handled for you, so from a Tekkotsu perspective users aren't | >expected to be needing to see that code. However, if you're looking to see some | >more complicated/intricate/fancy usage, then it might help. | > | >(more on Tekkotsu here if you haven't already: http://www.tekkotsu.org ) | > | >-ethan | > | (The article that 2005/09/22-05:58 posted, was modified by Author) | |
| |- 1858 RE:looking for commented sample code (OPEN-R ERS7) ejt [2005/09/22-08:08] |
| | Response : HomePage : Edit : Delete |
|
| >I
guess I should have mentioned that I am trying to get my university
started | >down the RoboCup path. We are a small team, and at this point we only have the | >manpower to work on one aspect of the code while using code from elsewhere to | >control other functions. | | A wise idea. Luckily there are a number of RoboCup teams to choose from, which | would get you off to a quick start. | | Regarding Tekkotsu, it's a good choice for general purpose robotics, but there's | still some ground to be covered if you are looking to use it for RoboSoccer, | particularly regarding localization, which we don't provide out-of-the-box. | However, this past year we had our first Tekkotsu-based Robosoccer teams, so | that may be changing. (I don't think they've hit the deadline for public release | yet, but I could put you in touch with them) | | On the other hand, if you're interested in using Tekkotsu's architectural | strengths, but not writing the soccer-specific stuff from scratch, it's often | not too hard to combine modules from other Robosoccer teams. For instance we | started out with the walk engine from CMU's 2002 robosoccer entry, but recently | after seeing a UPenn demo I spent a couple hours to port in the walk engine from | their 2004 entry so that can be used as well. Similarly, you can take | best-of-breed components from teams for other components you don't want to | tackle from scratch. | | Or you can just take an established team's code altogether as your basis, but of | course there's sometimes some politicial/support issues with that ;) | |
| |- 1866 RE:looking for commented sample code (OPEN-R ERS7) outsider [2005/10/04-00:07] |
| | | Response : Edit : Delete |
|
| | Hi, | | I wrote myself some useful material for learning OPEN-R (tutorials, | | presentation, code examples...) and used it on a summer course on Aibo | | programming. You can find all that material at http://www.ouroboros.org | | | | Hope it helps someone. | | | | Ricardo | | |
| | +- 1867 RE:looking for commented sample code (OPEN-R ERS7) chicoisl [2005/10/04-07:21] |
| | Response : Edit : Delete |
|
| Thanks,
I'll look into it. | | >Hi, | >I wrote myself some useful material for learning OPEN-R (tutorials, | >presentation, code examples...) and used it on a summer course on Aibo | >programming. You can find all that material at http://www.ouroboros.org | > | >Hope it helps someone. | > | >Ricardo | |
| +- 1880 RE:looking for commented sample code (OPEN-R ERS7) jcbaillie [2005/10/18-16:48] |
| Response : HomePage : Edit : Delete |
|
Hi, You mention that the URBI tutorial is for ERS200 but... ? it's not. URBI works with ERS7 and the tutorial covers that part. I'm puzzled :) Anyway, there is a new tutorial on the way which will be much easier and comprehensive than the current one. URBI is a good choice for robocup because it is very fast to code with it and your code will typically be easier to maintain. Take 5 min and you will be ready to run with URBI. Tekkotsu is a good choice too, but you must invest some time to understand it. Also, URBI is multi robot and benefit from the development of modules from all over the world, including companies and universities. You can use URBI with Java, Matlab or C++, on Windows, Linux or Mac OSX. Let me know if you are interested and I can give you a more detailed introduction. http://www.urbiforge.com Good luck! jcbaillie >I guess I should have mentioned that I am trying to get my university started down the RoboCup path. >We are a small team, and at this point we only have the manpower to work on one aspect of the code >while using code from elsewhere to control other functions. For that reason, compatibility with other >code is important. Would we be limiting ourselves in that respect with Tekkotsu (or URBI)? > >Also, I noticed that jcbaille's tutorial is for an ERS200 AIBO. It's quite helpful, but the code no longer >matches up with the ERS7 code. Is there anything else out there? > >Thanks for your help so far, >-Alex > >> >> >>This is probably overkill for a request for "sample" code, but you can peruse >>the Aperios-specific portion of Tekkotsu here: >> http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/aperios/ >>The "Capitalized" names are Aperios OObjects -- the others are 3rd party >>libraries I've ported to Aperios. >> >>Of course, the whole point of Tekkotsu is the complexity of the Aperios >>interaction is handled for you, so from a Tekkotsu perspective users aren't >>expected to be needing to see that code. However, if you're looking to see some >>more complicated/intricate/fancy usage, then it might help. >> >>(more on Tekkotsu here if you haven't already: http://www.tekkotsu.org ) >> >>-ethan >> >(The article that 2005/09/22-05:58 posted, was modified by Author) |
| V 1879 Set MCI device for CATPDirectSound dbroomfield [2005/10/17-06:29] |
| Response : Edit : Delete |
|
I
am trying to capture AIBO audio to a secondry audio device. It appears
that you need to use Sony's CATPDirectSound header to capture the Audio data but it doesnt appear to exponse the directsound enumerate devices property.. Anyone got any ideas? |
| V 1878 Convert GetAudioData buffer to waveOutWrite buffer dbroomfield [2005/10/14-05:36] |
| Response : Edit : Delete |
|
Can
someone help me.. I am no expert when it comes to C++ and I would like
to playback sound using a method other than directsound. The standard vc++ media libaries use waveOutWrite to write sound to the sound buffer.. I can work this but the RFW samples output the data as an unsigned char and this command needs a LPSTR. Do I need to perform some sort of conversion for this to work? |
| V 1871 AUDIO Streaming dbroomfield [2005/10/11-01:18] |
| | Response : Edit : Delete |
|
| From
what I can gather to send/receive audio you need to write them to a | staging wav file and ftp the file to AIBO. | | Is there a way to stream the audio data directly? | |
| +- 1873 RE:AUDIO Streaming AiboPet [2005/10/11-01:39] |
| | Response : HomePage : Edit : Delete |
|
| >From
what I can gather to send/receive audio you need to write them to a | >staging wav file and ftp the file to AIBO. | >Is there a way to stream the audio data directly? | | I assume you are using RFW. Both are supported (for playing audio, capturing | the AIBO microphone data is different) | | You can send a "staged" WAV file to AIBO and have it play | with "AudioSendWAVFile". | You can stream live audio samples using "AudioSendData". Timing is important | (buffering on AIBO is minimal) | There is also an explicit FTP upload function. You need to use it directly. | | The Audio system to be setup properly (the init code is tricky - and different | depending on the approach used). Copy the init code from one of the working | samples that is close to what you are trying | |
| +- 1874 RE:AUDIO Streaming AiboPet [2005/10/11-01:40] |
| | Response : HomePage : Edit : Delete |
|
| CORRECTION: | >There is also an explicit FTP upload function. ... | You do NOT need to use it directly. | |
| +- 1875 RE:AUDIO Streaming dbroomfield [2005/10/11-17:18] |
| | Response : Edit : Delete |
|
| Thanks
Pet. | | I actualy went though the AudioIn sample and that does most of what I want | although I am not an MFC man so I do struggle with the examples. | | What I am actually working on is trying to capture AUDIO from a given Audio | device in the control panel and send the data to AIBO.. I think the RemoteTest | is the only sample that does two/way AUDIO so it may take me some time to | disect that one! | |
| +- 1877 RE:AUDIO Streaming dbroomfield [2005/10/12-04:35] |
| Response : Edit : Delete |
|
ok
next question. I have taken a good look at the audioget sample and that uses directsound.. does anyone have a sample code to process audiogetdata() without using directsound.. preferably to windows waveout? |
| V 1876 Strange ImageOutputCaptureData behaviour aibo_user [2005/10/12-02:47] |
| Response : HomePage : Email : Edit : Delete |
|
Hi to all! We are 2 programmers with a *strange* problem with RFW... Working vith Visual C++ 2005 Express (but also 2003) the ImageOutputCaptureData function has different behaviours depending to the type of application: In native Win32 applications it works perfectly In .NET CLR Windows Forms ones, it doesn't. Notes: 1) The source code using for handling CVAIBO is *the same* in both cases 2) Video initialization is made correctly (to ensure this, we check ImageTP status, and it's IMAGETP_STATE_PLAY. 3) In both cases, the function returns correctly 0, without errors. 4) Other functions like Connect-Disconnect, ExternalControl, DoWalking... work well in both cases. What's happening? |
| V 1868 RCODE TELNET troubleshoot problem swei80 [2005/10/07-11:47] |
| | Response : Email : Edit : Delete |
|
| Hi,
I just connect to my AIBO thru Telnet. However, how do I trace the error | in my CODE??? I try using "RUN" statement, but it just result in a "STOP" | command. | | I put VDUMP, PRINT statement inside but nothing seems to come out???? | Can anyone give advice on this??? | | With Regards~ | Swei80 | |
| +- 1869 RE:RCODE TELNET troubleshoot problem AiboPet [2005/10/08-00:36] |
| | Response : HomePage : Edit : Delete |
|
| "PRINT" should work, try: | | PRINT "Hi There" | PRINT "%d" Pink_Ball | | Be sure to type it exactly (typos, punctuation and incorrect case can cause | problems) | ----- | Tracing is minimally supported | Try: | SET Trace 1 | | Then RUN. You can set other "Trace" values. | | SET Trace -1 | Will print a lot more info. | |
| +- 1870 RE:RCODE TELNET troubleshoot problem swei80 [2005/10/10-01:30] |
| | Response : Email : Edit : Delete |
|
| Hi,
I think the PRINT only will work when your program has no error. My | error.log has 13 CAN'T FIND IF STATEMENT errors. I walkthrough my code, and | even with my friend checking for me (to prevent ego programming), still we | couldn't locate the error. I try running with a R-CODE.R that I set with the | YART problem and i could get the PRINT msg. However on the code i modify, it | simply display the error for me and nothing else. | | I used the existing YART problem and modify the R-CODE.R. I couldn't locate | the error where the error log points. I mean if i misplaced a IF or ENDIF | statement, its quite reasonable but 13 same errors? hmmmmm.... | | What do you think, bro? | btw thanks for all your response | >"PRINT" should work, try: | > | >PRINT "Hi There" | >PRINT "%d" Pink_Ball | > | >Be sure to type it exactly (typos, punctuation and incorrect case can cause | >problems) | >----- | >Tracing is minimally supported | >Try: | >SET Trace 1 | > | >Then RUN. You can set other "Trace" values. | > | >SET Trace -1 | >Will print a lot more info. | |
| +- 1872 RE:RCODE TELNET troubleshoot problem AiboPet [2005/10/11-01:28] |
| Response : HomePage : Edit : Delete |
|
>
My error.log has 13 CAN'T FIND IF STATEMENT errors. Well there's your problem ^^^ Fix that first. The program will not run if there are basic syntax errors (in ERROR.LOG). Things like mis-matched IF/THEN/ELSE/ENDIF or WHILE loops are common when you are writing code from scratch (especially if new to RCODE) The RCODE system is an interpreter and a runtime. It is not a fullblown compiler. Its error reporting is very poor [which is why you must be very very careful when typing in the code]. John of Dogsbody wrote some simple parsing/checking tools - but IMHO once you learn the rules, you don't need them if you are careful. > I mean if i misplaced a IF or ENDIF >statement, its quite reasonable but 13 same errors? hmmmmm.... Look at it again (or go back and start with one of your IF statement - and see why it isn't working compared to the others in the .R file) You are probably assuming something about the syntax that isn't supported. Copy working code (from the default YART RCODE.R file). When nesting IF/ELSE blocks the rules are very strict. |
| V 1862 New AIBO model ERS-7m3, will RFW work right away? wolftag [2005/09/30-07:13] |
| | Response : Edit : Delete |
|
| I
finally got my boss to approve purchasing an AIBO, and now they are all
out | of stock! Does anyone know if the new model will be immediately supported by | the various AIBO SDE's? | |
| +- 1865 RE:New AIBO model ERS-7m3, will RFW work right away? AiboPet [2005/10/02-01:35] |
| Response : HomePage : Edit : Delete |
|
>I
finally got my boss to approve purchasing an AIBO, and now they are all
out >of stock! Does anyone know if the new model will be immediately supported by >the various AIBO SDE's? Based on past history, I suspect there will be an update with the RFW DLLs with the new AiboEntertainmentPlayer 2.0. Historically this comes out shortly after the robots are available. When the old RFW was relaced by the Mind2 version (about 1 year ago), there were minor API changes too. |
| V 1863 Strange sitting + fall down condition aibo_user [2005/09/30-22:16] |
| | Response : HomePage : Email : Edit : Delete |
|
| Bug? Coma? Devil's possession? Mistery... | | My AIBO has sometimes a strange behaviour... | 1) It sits in a strange position (like the classic one, but with the rear legs | straight) | 2) Fall down condition (front legs twirling) | 3) Aibo restores from fall down condition (tries to lie) | 4) Aibo reaches stand position | 5) Aibo freezes. Only sensor lights, if touched, react. | | What's happening to it (or him)? | |
| +- 1864 RE:Strange sitting + fall down condition AiboPet [2005/10/02-01:32] |
| Response : HomePage : Edit : Delete |
|
>Bug?
Coma? Devil's possession? Mistery... > >My AIBO has sometimes a strange behaviour... I suggest asking on a user BBS, like AiboLife http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi This BBS is focused on SDE development topics. Quick advice: try taking out the memory stick and boot the dog in clinic / self-test mode. Read the manual to be sure it works the way it should. |
| V 1861 [Remote Framework] buffers to WAV ferrante [2005/09/30-02:28] |
| Response : HomePage : Email : Edit : Delete |
|
A programming question... Has anyone a *simple* code snippet to record audio from Aibo into wav? Or... to grab data from one or more buffers (with AudioGetData) and put them into a *hearable* wav file? |
| V 1860 ImageGetFrameRate always returns 0 ccbradub [2005/09/27-01:57] |
| Response : Email : Edit : Delete |
|
Hi! I am trying to get the image rate of my AIBO using the function from RFW, but always I get value 0. Actually (and here I am a little bit confused), does ImageGetFrameRate return the status of the function call (as far as I could understand from the documentation) or return the effective frame rate of the camera? + ImageGetFrameRate [Function] Get the frame rate of image transmission [Return] Success: VAIBO_NOERROR Fail: Error code (see Error code list at the top of this text) [Description] Get the frame rate of image transmission Is the frame rate information embedded in another parameter/structure? Thanks! Bogdan |
| V 1855 odataFBK_YCrCb_HPF pakitoMR [2005/09/21-07:14] |
| | Response : HomePage : Email : Edit : Delete |
|
| In
the ImageObserver example the image type captured can be odataFBK_YCrCb
or | odataFBK_YCrCb_HPF. The odataFBK_YCrCb_HPF type image is a double size image | with a high pass filter. My camera takes images in this type, but I have not | selected it. How can I select between odataFBK_YCrCb_HPF or odataFBK_YCrCb? Does | this type depends only on the hardware? In other words, is it selected | automatically depending on the camera? I haven't found any information in the | Open-R docs or in the web.... | |
| +- 1856 RE:odataFBK_YCrCb_HPF AiboPet [2005/09/22-02:08] |
| Response : HomePage : Edit : Delete |
|
>
How can I select between odataFBK_YCrCb_HPF or odataFBK_YCrCb? Does >this type depends only on the hardware? Depends on the version of OpenR SDK system software. All current releases should send 6 band "HPF" image data (2 color bands, 4 Y bands - 352x288 for ERS-2x0/31x or 416x320 for the ERS-7) Older versions of the OpenR SDK had only 3 bands of image data (2 color One Y band - 176x144 lores). The "ReconstructImage" technique didn't work back then. The 6 band "HPF" format is backwardly compatible. Ignore the last 3 bands and it is same as the lores option. |
| V 1847 Do you have OPEN-R built-in recognize the program of music? abc [2005/09/16-19:42] |
| Response : Email : Edit : Delete |
|
Do
you have OPEN-R built-in recognize the program of music? for ers-7. (The article that 2005/09/13-22:06 posted, was modified by Author) |
| V 1848 Linux Paths in Cygwin gj15987 [2005/09/14-23:58] |
| | Response : Email : Edit : Delete |
|
| I am
trying to install gcc using the cygwin...erm..."thing". Sorry not too | clued up on this yet. | | I'm going through the installation guide and i've installed cygwin. I'm now | onto installing gcc with it. I'm at the stage where it says this | | cd /usr/local | tar zxvf /xxx/mipsel-devtools-3.3.2-bin-r1.tar.gz | | with /xxx being the directory that the file is stored in. However i've tired | all sorts of /user/local and /home/Grant etc but it keeps saying file or | directory not found. | | Cygwin is installed in C:\cgywin and the file i'm trying to install is in the | same folder. | | Can anyone tell me what the /xxx is supposed to be?! | | Thanks Very Much. | | Grant. | |
| +- 1849 RE:Linux Paths in Cygwin AiboPet [2005/09/15-00:20] |
| | Response : HomePage : Edit : Delete |
|
| >cd /usr/local | >tar zxvf /xxx/mipsel-devtools-3.3.2-bin-r1.tar.gz | | If you are running within, the cygwin shell, then "cd /usr/local" should work. | You should run tar and everything else within a cygwin shell (click on | the "Cygwin" icon installed on your desktop - then type in the commands) | | --- | The real directory is "C:\CYGWIN\usr\local". | NOTE: "usr" not "user" | |
| +- 1850 RE:Linux Paths in Cygwin gj15987 [2005/09/15-07:28] |
| Response : Email : Edit : Delete |
|
Ah,
i sussed it. Thanks. (The article that 2005/09/15-00:42 posted, was modified by Author) |
| V 1835 Problem of open-r program? abc [2005/09/10-18:44] |
| | Response : Email : Edit : Delete |
|
| HOW
to modify OPEN-R PROGRAM TO LET MUSIC TRIGGER to start AIBO IN THE short | TIME ? | FOR ERS-7. | |
| +- 1837 RE:Problem of open-r program? AiboPet [2005/09/10-23:57] |
| | Response : HomePage : Edit : Delete |
|
| >HOW
to modify OPEN-R PROGRAM TO LET MUSIC TRIGGER to start AIBO IN THE short | >TIME ? | >FOR ERS-7. | | The Open-R SDK does *not* have music detection/triggering built in. You must | write your own frequency detection algorithm to determine which musical notes | are heard by AIBO, and compare them with a list of known trigger songs. | I don't know of an open source version specific for the AIBO. | | --- | As mentioned earlier, RFW and RCODE *do* have music triggering software built | in, but you are limited to fixed trigger songs. | I still recommend using RCODE if you can live with this limitation. | | I may be able to give better advice if you give more details on what you are | trying to do. | |
| +- 1838 RE:Problem of open-r program? abc [2005/09/11-01:18] |
| | Response : Email : Edit : Delete |
|
| I
want to let the aibo listen to music trigger and then start in action
in a | short time, then music end he can come after to stop. | Excuse me, what method you have to complete a top what I describe. | (The article that 2005/09/11-01:13 posted, was modified by Author) | |
| +- 1846 RE:Problem of open-r program? AiboPet [2005/09/13-00:32] |
| Response : HomePage : Edit : Delete |
|
>I
want to let the aibo listen to music trigger and then start in action
in a >short time That's what Disco AIBO does. The RCODE program is very simple. http://www.aibohack.com/disco/index.html#ERS7 > , then music end he can come after to stop. The trigger starts the movement. It will stop after the action is complete (the canned dance in the case of Disco AIBO). |
| V 1845 OPEN-R SDK build fails on FC4 androticus [2005/09/12-15:05] |
| Response : Email : Edit : Delete |
|
I
have i386 Fedora Core 4 (up to date) running on an Opeteron 64. When I try to build the devtools using the build-devtools script, there are many fatal errors. I started looking them up -- the first one, an issue with obstack.h could be fixed by replacing the file. But then I got additional errors. It looks like this gcc cannot be built properly with gcc 4 in FC4. Has anyone successfully built OPEN-R mipsel tools on FC4? Will Sony be updating the SDK for gcc 4? first failure: ../../gcc-3.3.2/gcc/read-rtl.c: In function read_rtx: ../../gcc-3.3.2/gcc/read-rtl.c:653: error: invalid lvalue in increment make[1]: *** [read-rtl.o] Error 1 make[1]: Leaving directory /usr/local/OPEN_R_SDK/bld-gcc-3.3.2/gcc make: *** [all-gcc] Error 2 second failure: ../../gcc-3.3.2/gcc/cp/decl.c: In function 'push_binding_level': ../../gcc-3.3.2/gcc/cp/decl.c:727: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'pop_binding_level': ../../gcc-3.3.2/gcc/cp/decl.c:783: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'suspend_binding_level': ../../gcc-3.3.2/gcc/cp/decl.c:800: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c:819: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'resume_binding_level': ../../gcc-3.3.2/gcc/cp/decl.c:832: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'pushdecl_with_scope': ../../gcc-3.3.2/gcc/cp/decl.c:4544: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c:4546: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'cxx_init_decl_processing': ../../gcc-3.3.2/gcc/cp/decl.c:6945: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'grokdeclarator': ../../gcc-3.3.2/gcc/cp/decl.c:10791: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c:10794: error: invalid lvalue in assignment ../../gcc-3.3.2/gcc/cp/decl.c: In function 'start_function': ../../gcc-3.3.2/gcc/cp/decl.c:14434: error: invalid lvalue in assignment Any help is appreciated. (The article that 2005/09/12-14:19 posted, was modified by Author) |
| V 1771 gcc update II ejt [2005/06/30-03:00] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
posted before (msg 1662) a request to update to 3.4 for performance
reasons, | but now I have a technical reason. I want to implement a clone() operation so I | can make copies of polymorphic objects. | | The straightforward implementation is: | class Base { | public: | virtual Base* clone() const; | }; | | class Derived : public Base { | public: | virtual Derived* clone() const; | }; | | notice the return types -- each returns its own type. This is legal C++, but | unsupported in g++ 3.3.x and prior: | "sorry, unimplemented: adjusting pointers for covariant returns" | However, it *is* supported in 3.4 | | Is anyone planning to update gcc ever again? Pretty pretty please? | | thanks, | -ethan | |
| |- 1792 RE:gcc update II openrsupport@(Administrator) [2005/07/13-01:53] |
| | Response : Edit : Delete |
|
| Ethan, | | Thank you for supporting the AIBO SDE community. Unfortunately, we don't have | any current plan to support gcc 3.4. We will inform you if this changes. | | AIBO SDE Support | |
| +- 1832 RE:gcc update II nlopes [2005/09/10-04:21] |
| | Response : Edit : Delete |
|
| Hi Ethan, | | Had you been able to build gcc 3.4 successfully? | I've tried to use the lastest versions of newlib and binutils, along with gcc | 3.4.4 and gcc 4.0.1, but I had no luck. I've modified the patches acordingly, | but I always get a compile error in gcc (binutils builds fine). | | My last script attempt: | http://mega.ist.utl.pt/~ncpl/build-devtools-3.4.4.sh | uses: | * binutils-2.16.1.tar.bz2 | * gcc-3.4.4.tar.bz2 | * newlib-1.13.0.tar.gz | | Nuno | |
| +- 1834 RE:gcc update II ejt [2005/09/10-13:02] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Had
you been able to build gcc 3.4 successfully? | No, I tried once but wound up giving up. I'm willing to give it another shot though... | | Any particular reason you're shooting for 3.4.4 instead of 4.0.1? I'm not much up on the reputation 4.x | has been getting, though I've been doing a lot of development with the 4-series branch Apple provides | with their system without any trouble. | | -ethan | |
| +- 1836 RE:gcc update II nlopes [2005/09/10-19:30] |
| | Response : Edit : Delete |
|
| >Any
particular reason you're shooting for 3.4.4 instead of 4.0.1? I'm not | much up on the reputation 4.x | | I've tried both. I'll use one of them, just need it to work :) | gcc 4.0.1 should be stable enough for us (when gnu GMP guys say that a | compiler is good, you can trust..) | |
| |- 1839 RE:gcc update II ejt [2005/09/11-04:09] |
| | Response : HomePage : Email : Edit : Delete |
|
| Sounds
good -- I'm trying to get 4.0.1 working now... run into trouble with
some "unwind" files, looks | like perhaps 'inhibit_libc' should be defined some places but isn't? I'm still reading up on how these | pieces fit together. Looks like generally gcc assumes libc, which newlib replaces, so 'inhibit_libc' | should be off (since newlib is supposed to be used instead, so 'libc' is enabled, it's just not the *same* | libc). Perhaps gcc expects some functionality from libc which newlib doesn't actually provide? | | For now I'm just adding a '#define inhibit_libc' in only the sources which complain about something | missing, works so far, but haven't gotten to an executable yet... newlib has a mailing list which might | be able to help us out if it doesn't work. | | -ethan | |
| +- 1840 RE:gcc update II ejt [2005/09/11-13:51] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
got through binutils and gcc -- actually was able to compile Tekkotsu
with it (and turned up a few | new warnings for things to fix), but ran into trouble with the linking, realized newlib never actually | installed... newlib's configure runs through, seems to not find any of the mipsel compilers, but | completes successfully, so make triggers, but doesn't do anything, and so some required libraries (e.g. | libm) wind up missing. But at least the compiler seems to be working ;) | | I'll post the build script I'm using after I play with newlib a bit more. (hoping you might have a | suggestion on that one) | | -ethan | |
| +- 1841 RE:gcc update II ejt [2005/09/12-04:03] |
| | Response : HomePage : Email : Edit : Delete |
|
| OK,
looks like things are almost done. I've posted my modified build script
at: | http://ejt.no-ip.org/stuff/build-devtools-4.0.1.sh | | It's a little smarter in handling inputs -- checks for either .bz2 or .gz files, only untars if the directory | doesn't already exist, only patches if it untars, etc. so you can run the script multiple times without it | trying to rebuild from scratch every time. | | You also need to patch OPEN_R_SDK/OPENR/bin/mkbin to handle a change in gcc's command line | interface: | http://ejt.no-ip.org/stuff/mkbin.diff | Otherwise you get errors because collect2 can't be found. | | Currently, the sticking point is std::__default_alloc_template seems to be undefined. | Everything works up to the very end, where we get a lot of these: | /usr/local/OPEN_R_SDK_new/OPEN_R/lib/libObjectComm.a(ShmUnitAllocator.o): In function | `ShmUnitAllocator::ShmUnitAllocator()': | ShmUnitAllocator.cc:(.text+0x30): undefined reference to `std::__default_alloc_template<true, | 0>::allocate(unsigned int)' | ShmUnitAllocator.cc:(.text+0x58): undefined reference to `std::__default_alloc_template<true, | 0>::allocate(unsigned int)' | ... | | I think if we fix this, we're golden. Looks like this is supposed to be supplied by libstdc++? Any | suggestions? | | -ethan | (The article that 2005/09/11-18:39 posted, was modified by Author) | |
| +- 1842 RE:gcc update II ejt [2005/09/12-02:18] |
| | Response : HomePage : Email : Edit : Delete |
|
| OK,
it looks like: | std::__default_alloc_template<bool,int> | was replaced by a slightly different class in 3.4+: | __gnu_cxx::__pool_alloc<T> | as reported here: | http://gcc.gnu.org/onlinedocs/libstdc++/20_util/allocator.html | | Which brings us to the final problem of binary compatability -- I could write an adapter to forward calls | to the new __pool_alloc, but the real problem is that code compiled with 3.3 can't be linked against | code compiled with 3.4 (or beyond). So now the ball is in Sony's court -- they need to recompile the | SDK with 4.0.1 in order to let us use it. I'll start a new thread to get their attention. | | -ethan | |
| +- 1843 RE:gcc update II nlopes [2005/09/12-02:55] |
| Response : Edit : Delete |
|
Very
nice work, Ethan ;) Sorry I couldn't help you, but I only arrived home now. then I see your posts here and in the newlib ML.. very good work!! >Which brings us to the final problem of binary compatability -- I could write an adapter to forward calls >to the new __pool_alloc, but the real problem is that code compiled with 3.3 can't be linked against >code compiled with 3.4 (or beyond). So now the ball is in Sony's court -- they need to recompile the >SDK with 4.0.1 in order to let us use it. I'll start a new thread to get their attention. Good idea. I hope they'll do this, so that our aibos start to fly :) |
| V 1833 URBI 0.9.6 is available jcbaillie [2005/09/10-07:47] |
| Response : HomePage : Email : Edit : Delete |
|
URBI version 0.9.6 is now available on urbiforge: http://www.urbiforge.com URBI (Universal Robotic Body Interface) is a new "interface language" used to control robots in general and Aibo in particular. It is based on a client/server architecture and allows interesting event driven parallel processing of commands, easy motor/sensor access and several features adapted for robotics. More importantly, URBI is designed with simplicity in mind and can be used in a couple of minutes. The ball tracking head program is 3 lines long only (in a C-like syntax). It is very easy to write interesting action/perception loops in URBI or even complex behaviors (also, native behavior description support will be included in the language soon). URBI can be interfaced with C++, Java or Matlab using the URBI Library. The idea is that you can use URBI and program your Aibo with your favorite language, on Windows, Linux or MacSOX as you like. More can be found on the "What is URBI" page and the FAQ: http://www.urbiforge.com/eng/whatisurbi.html http://www.urbiforge.com/eng/faq.html URBI has been used by Diego Pardo to write the Aibo dance that won the "Daft Punk" contest; it is used by academic research labs and aibo hobbyists programmers as well. We are very much interested in feedback from the Aibo community. Don't hesitate to use the forum on urbiforge to ask questions, request features or submit URBI programs for the community to share. Thanks, Best Regards, JC Baillie |
| V 1829 AIBO falls asleep with examples from tutorials - why? xander [2005/09/01-20:54] |
| | Response : Email : Edit : Delete |
|
| I am
studying the masterpiece robots AIBO, especially I am trying to learn to | program them. I have downloaded lots of documents from AIBO web site | (https://openr.aibo.com/openr/eng/perm/download.php4) and have read some of them. | | I have followed the instructions in InstallationGuide_E.pdf (from | OPEN_R_SDK-docE-1.1.5-r1.tar.gz), I mean the ones in point 3.2.1. Everything | word for word up to point 4 because when I booted the AIBO it woke for a while | and then fell asleep again. I do not understand where the problem is. In some | other document (SDK_Tutorial1Web_E.pdf) there were similar instructions. Here | are two things I have noted that bother me: | - first, the 'original' memory stick has an extra directory (APP) | - second, the files, which the programmer should edit to determine which object | to be executed (like ones in MW/CONF - CONNECT.CFG, DESIGN.CFG, DESIGNDB.CFG and | OBJECT.CFG in the examples) are in text, i.e. understandable format, whereas the | same files in the 'original' memory stick are binary | | How to keep the AIBO aweken and make it execute the new .BINs? | (The article that 2005/09/01-20:51 posted, was modified by Author) | |
| +- 1830 RE:AIBO falls asleep with examples from tutorials - why? AiboPet [2005/09/02-00:31] |
| | Response : HomePage : Edit : Delete |
|
| >
.. I booted the AIBO it woke for a while | >and then fell asleep again. | | Did the robot stretch his legs ? | If no, then I strongly suspect you are not using a "PMS" stick. | http://www.aibohack.com/pms.htm | | You need a PMS stick to run RCODE, RCodePlus or OpenR SDK programs ! | | ----- | >- first, the 'original' memory stick has an extra directory (APP) | That's a commercial memory stick/software (like AiboLife or AiboMind) - not | a "PMS" | It won't work for OpenR SDK programs (commercial software is almost all binary | encrypted, OpenR SDK programs are partially encrypted with a few plain text | files including your own) | |
| +- 1831 RE:AIBO falls asleep with examples from tutorials - why? xander [2005/09/02-17:07] |
| Response : Email : Edit : Delete |
|
>>
.. I booted the AIBO it woke for a while >>and then fell asleep again. > >Did the robot stretch his legs ? >If no, then I strongly suspect you are not using a "PMS" stick. >http://www.aibohack.com/pms.htm > >You need a PMS stick to run RCODE, RCodePlus or OpenR SDK programs ! > Thank you very much, this was the cause of the problem! |
| V 1825 Is there software to edit the LED BLINK ? abc [2005/08/29-18:48] |
| | Response : Email : Edit : Delete |
|
| Is
there software to edit the LED BLINK and match with the AIBO MOTION
EDITOR? | I use ers-7. | |
| +- 1826 RE:Is there software to edit the LED BLINK ? AiboPet [2005/08/30-00:27] |
| | Response : HomePage : Edit : Delete |
|
| >Is
there software to edit the LED BLINK and match with the AIBO MOTION
EDITOR? | >I use ers-7. | | Use Skitter | http://www.dogsbodynet.com/skitter.html | | It will edit .LED files (As well as .MTN files and others) | |
| +- 1827 RE:Is there software to edit the LED BLINK ? abc [2005/08/30-13:33] |
| | Response : Email : Edit : Delete |
|
| Before
i use skitter and it can't store file in the sony memory card . | What reason is that? | What is the difference of sony medit for ers-7 and skitter? | |
| +- 1828 RE:Is there software to edit the LED BLINK ? AiboPet [2005/09/01-00:45] |
| Response : Edit : Delete |
|
Also
check the FAQ: https://openr.aibo.com/openr/eng/perm/medit/regi_faq3.php4#C0007 >Before i use skitter and it can't store file in the sony memory card . >What reason is that? Skitter has not been updated to work directly with RCODE ERS-7 If using MEdit for motions use it with RTool for building ODA files, including LED files. See the instructions for MEdit and RTool. >What is the difference of sony medit for ers-7 and skitter? MEdit is the official editor provided with the SDK Skitter is a different tool written by an AIBO Owner FWIW: I prefer Skitter. I also recommend YART for getting started with RCODE programming, and YART_APG for combining WAVs, MTNs and LEDs (not RTool). However if you are already using MEdit and RTool, you should probably stick with them. |
| V 1824 How write LED blink program (*.Led file) in the R-CODE? abc [2005/08/28-02:42] |
| Response : Email : Edit : Delete |
|
How write LED blink program (*.Led file) in the R-CODE? Is it all right to give me an example file? I use ers-7. |
| V 1823 How let music trigger action and can't produce to delay? abc [2005/08/27-22:47] |
| Response : Email : Edit : Delete |
|
I use is R-CODE. I use different trigger music to trigger different action respectively . How combine them and can't produce to delay? If can combine them and can't produce to delay. How write its program? (The article that 2005/08/27-22:43 posted, was modified by Author) |
| V 1819 Can I use random MID music file to trigger ers-7? abc [2005/08/21-01:39] |
| | Response : Email : Edit : Delete |
|
| Can
I use random MID music file to trigger ers-7?Is it all right to proceed | many times trigger and don't delay in a music ?How does that write above- | mentioned 2 itemses program ? | (The article that 2005/08/21-01:36 posted, was modified by Author) | |
| +- 1820 RE:Can I use random MID music file to trigger ers-7? AiboPet [2005/08/21-00:17] |
| | Response : HomePage : Edit : Delete |
|
| >Can
I use random MID music file to trigger ers-7?Is it all right to proceed | >many times trigger and don't delay in a music ?How does that write above- | >mentioned 2 itemses program ? | | Please provide more information if you can. What tool are you using? RCODE? | OpenR-SDK ? RFW ? | | ----- | (another attempted reply) | You can trigger RCODE using certain MIDI sequences ("trigger sounds"). The | music is played on your computer and heard by the AIBO. | See the "AiboSound" and "AiboSound_ID" variables and the .MID files in | the "Redist7/AiboSound" folder of the RCODE SDK. | | You can insert the trigger sounds into your own music (like DiscoAibo). | | You can not change these trigger sounds. | | There are two sets of trigger sounds. The standard one in RCODE ERS-7 and a | special version for Disco Aibo. | |
| +- 1821 RE:Can I use random MID music file to trigger ers-7? abc [2005/08/21-02:33] |
| | Response : Email : Edit : Delete |
|
| The
tool that I use is RCODE. | Before I having already used (R-CODE SDK main archive for ERS-7\Redist7_ver2\ | aibosuond and aibotone) to trigger music. | Can I use in addition to( The standard one in RCODE and ERS-7 a special | version for Disco Aibo) Of trigger music? | If can. How write it to trigger musical program? | (The article that 2005/08/21-01:59 posted, was modified by Author) | |
| +- 1822 RE:Can I use random MID music file to trigger ers-7? AiboPet [2005/08/23-00:21] |
| Response : HomePage : Edit : Delete |
|
>Can
I use in addition to( The standard one in RCODE and ERS-7 a special >version for Disco Aibo) Of trigger music? Sorry, no. The necessary files are encrypted, and can only be changed by Sony. |
| V 1818 Author DELETE ***** [2005/08/20-16:41] |
| . |
|
(The
article that 2005/08/20-01:08 posted, was modified by Author) |
| V 1813 Problem with running RCODE swei80 [2005/08/16-10:49] |
| | Response : Email : Edit : Delete |
|
| Hi,
I had some problem running RCODE on my ERS-7. I done everything as
mention | in the tutorial as in unzipping the Redist7_ver1.zip and the related files | copied inside. But upon starting up, My AIBO doesn't do anything at all. | | I even try the Maze code or simple code such as | :Start | PLAY:ACTION:STAND | WAIT | | However, AIBO doesn't stand according to my code. | Anyone here know where my problem lies???? | SOrry for asking a noob question~~~ | |
| +- 1817 RE:Problem with running RCODE AiboPet [2005/08/19-00:42] |
| Response : HomePage : Edit : Delete |
|
>Hi,
I had some problem running RCODE on my ERS-7... My general advice for people starting up is to run one of the pre-canned RCODE/YART personalities. For example, Cylon for the ERS-7 version http://www.aibohack.com/ers7/cylon7.htm First it is a good sanity test for your hardware and setup process (ie. you need a "PMS" stick, a stick reader - SONY BRAND, and the ability to install 2 ZIP files). If it doesn't work, you know there is something very basic going wrong. If you do want to use YART, it is immediately available for you to play with. If you don't want to use YART, it has all the regular RCODE (ERS-7) functionality [YART is simply an addon]. You can edit the existing R-CODE.R file, or replace it with anything else you like, including the .R files from the RCODE SDK (be sure to only use ERS-7 sample scripts) If the issue is the tutorial, please ask the person who wrote it. |
| V 1814 Can I build up not- built-in recognize MID file for ers-7? abc [2005/08/18-18:39] |
| | Response : Email : Edit : Delete |
|
| Can I build up not- built-in recognize MID file for ers-7? | (The article that 2005/08/18-18:34 posted, was modified by Author) | |
| +- 1816 RE:Can I build up not- built-in recognize MID file for ers-7? AiboPet [2005/08/19-00:35] |
| Response : HomePage : Edit : Delete |
|
>Can
I build up not- built-in recognize MID file for ers-7? I don't fully understand what you are asking. You can create your own .MID music files for playing on the ERS-7. This is extremely easy using RCODE, RFW and even tweeking AiboMind1/2 canned music. https://openr.aibo.com/openr/eng/perm/regi_faq.php4#A0008 (SDE FAQ question #8) There are some restrictions on the format (timing values etc). The ERS-7 will play polyphonic MID files. |
| V 1812 Where can I read the network status? olafus [2005/08/16-02:46] |
| | Response : Email : Edit : Delete |
|
| I
have a java application reading the image and now I want to read de
network | status but I don't find in which port it is transmitted. At least not in 59000. | | Can anyone help me? | (The article that 2005/08/16-02:45 posted, was modified by Author) | |
| +- 1815 RE:Where can I read the network status? AiboPet [2005/08/19-00:30] |
| Response : HomePage : Edit : Delete |
|
>I
have a java application reading the image and now I want to read de
network >status but I don't find in which port it is transmitted. At least not in 59000. Please provide more info. What software are you running on the AIBO ? What net protocol ? What model AIBO ? |
| V 1810 How to walk? moekel [2005/08/12-21:08] |
| | Response : Email : Edit : Delete |
|
| Hi | I'm just beginning in OPEN-R and I've tested the sample-programs on the ERS7. | Now I would like to let the ERS7 walk. | I don't want to use Tekkotsu - it's just overkill for me at this time. I just | need some joint-values to let the ERS7 move around (standing, not kneeing). | Does anybody have a complete sequence of joint-values or an idea to figure out? | Just post here or mail to aibo@moekel.de | Thanks! | | Antworten gern auch in Deutsch. | |
| +- 1811 RE:How to walk? AiboPet [2005/08/13-02:13] |
| Response : HomePage : Edit : Delete |
|
>Hi >I'm just beginning in OPEN-R and I've tested the sample-programs on the ERS7. >Now I would like to let the ERS7 walk. Unfortunately there is no Open-R ERS-7 component that does full walking (there is one for the ERS-210/220 used in SoccerLion) ---- >I don't want to use Tekkotsu - it's just overkill for me at this time. BTW: they use a tweeked version of the CMU walk engine That is also available in my "SpeedRacer" program (or earlier Walkbench). It uses straight Open-R: http://www.aibohack.com/openr_sdk/speedracer.htm http://www.aibohack.com/openr_sdk/SpeedRacer01_SRC.zip This is an algorihmic walk engine. It is a "knee" walk ======= > I just >need some joint-values to let the ERS7 move around (standing, not kneeing). >Does anybody have a complete sequence of joint-values or an idea to figure out? Check out the "MoNet" sample. It has a few sample walk motion files. The dog walks normally (ie. mostly standing, not a "knee walk"). It is not algorithmic. If you are happy with the canned walking styles, use it. |
| V 1797 R-CODE and BALL/BONE PERCEPTION matan325 [2005/07/20-21:59] |
| | Response : Email : Edit : Delete |
|
| I am
trying to get the AIBO (ers-7) behave using some simple R-CODE scripts. | Unfortunately it works very bad. I made a simple loop just to allow me to place | objects (Ball and Bone) in front of AIBO and see whether it detects them | (correctly) or not: | | DO | IF:Pink_Ball:==:0:THEN | IF:AIBONE:==:0:THEN | PLAY:MWCID:2397 //Nothing seen - Stilleye | ELSE | PLAY:MWCID:2394 //Bone seen - boneeyes! | SET:AIBONE:0 | ENDIF | ELSE | IF:AIBONE:==:0:THEN | PLAY:MWCID:2391 //Ball - bardoubleeyes | SET:Pink_Ball:0 | ELSE | PLAY:MWCID:2395 //Bone & Ball seen - crosseyes | SET:AIBONE:0 | SET:Pink_Ball:0 | ENDIF | ENDIF | LOOP | | This works very bad. Sometimes it's detected correctly of course but 50% of the | time it's not. Things are detected when there is nothing to be seen. Both bone | and ball is detected when only one of them is in front of the AIBO. Nothing is | detected when AIBO is obviously facing the object. Sometimes it just oscillates | between just about all the different states. This leads to my question: | | What are the optimal environment for the AIBO R-CODE vision routines to perform | well? Black background? White background? How much light? Ordinary room light a | rainy day? A lot of light? Quite-some-light-but-not-TOO-much? I really need to | get this straight. Perhaps I do something wrong with the R-CODE script? | Expecting the wrong things? | | Have a nice day! | | /Mats Andr$BqO(B | (The article that 2005/07/20-20:45 posted, was modified by Author) | |
| |- 1798 RE:R-CODE and BALL/BONE PERCEPTION AiboPet [2005/07/21-01:25] |
| | Response : HomePage : Edit : Delete |
|
| >I
am trying to get the AIBO (ers-7) behave using some simple R-CODE
scripts. | >Unfortunately it works very bad... | Several problems: | ---- | Issue #1) After playing an action, you should "WAIT" until it is completed | | PLAY:MWCID:2397 //Nothing seen - Stilleye | WAIT | | ---- | Issue #2) If you are going to clear the system variables yourself | (eg: "SET:AIBONE:0" be sure that you do it at the right time, and only when it | is set. | See the [BAD] comments added to your source code below. They are cases where | you detect no input, play something, then clear the input variable (even if it | was not set). This will cause the "50%" detection problem because the system | may set the variable in the middle (after you test it with IF, but before you | clear it with SET). The value is lost. | | ----- | Issue #3) after you get this fixed, you will find the LEDs will flash their | state for only a short time the first time it sees a ball/bone. You may want | the LEDs to stay on all the time the dog sees it and only turn off when you | remove the ball from his view. | To do that, store the last state in a variable and compare it with the current | state. Never set the "Pink_Ball" or "AIBONE" variables in the main loop. | See the skeleton R-CODE.R provided with YART and look for the "lastPink" | and "lastBone" variables and comparison logic. | http://www.aibohack.com/ers7/yart7.htm | ======= | | >DO | > IF:Pink_Ball:==:0:THEN | > IF:AIBONE:==:0:THEN | > PLAY:MWCID:2397 //Nothing seen - Stilleye | > ELSE | > PLAY:MWCID:2394 //Bone seen - boneeyes! | [BAD] > SET:AIBONE:0 | > ENDIF | > ELSE | > IF:AIBONE:==:0:THEN | > PLAY:MWCID:2391 //Ball - bardoubleeyes | [BAD] > SET:Pink_Ball:0 | > ELSE | > PLAY:MWCID:2395 //Bone & Ball seen - crosseyes | [possibly BAD] > SET:AIBONE:0 | [possibly BAD] > SET:Pink_Ball:0 | > ENDIF | > ENDIF | >LOOP | > | |
| +- 1809 RE:R-CODE and BALL/BONE PERCEPTION jcbaillie [2005/08/03-11:47] |
| Response : HomePage : Email : Edit : Delete |
|
Hi! I think you should have a look at URBI instead of RCODE: http://www.urbiforge.com I hope this helps ;) JC Baillie >I am trying to get the AIBO (ers-7) behave using some simple R-CODE scripts. >Unfortunately it works very bad. I made a simple loop just to allow me to place >objects (Ball and Bone) in front of AIBO and see whether it detects them >(correctly) or not: > >DO > IF:Pink_Ball:==:0:THEN > IF:AIBONE:==:0:THEN > PLAY:MWCID:2397 //Nothing seen - Stilleye > ELSE > PLAY:MWCID:2394 //Bone seen - boneeyes! > SET:AIBONE:0 > ENDIF > ELSE > IF:AIBONE:==:0:THEN > PLAY:MWCID:2391 //Ball - bardoubleeyes > SET:Pink_Ball:0 > ELSE > PLAY:MWCID:2395 //Bone & Ball seen - crosseyes > SET:AIBONE:0 > SET:Pink_Ball:0 > ENDIF > ENDIF >LOOP > >This works very bad. Sometimes it's detected correctly of course but 50% of the >time it's not. Things are detected when there is nothing to be seen. Both bone >and ball is detected when only one of them is in front of the AIBO. Nothing is >detected when AIBO is obviously facing the object. Sometimes it just oscillates >between just about all the different states. This leads to my question: > >What are the optimal environment for the AIBO R-CODE vision routines to perform >well? Black background? White background? How much light? Ordinary room light a >rainy day? A lot of light? Quite-some-light-but-not-TOO-much? I really need to >get this straight. Perhaps I do something wrong with the R-CODE script? >Expecting the wrong things? > >Have a nice day! > >/Mats Andr$BqO(B >(The article that 2005/07/20-20:45 posted, was modified by Author) |
| V 1805 Is it possible to recognize Orange Balls in R-CODE? JFCBowdoin [2005/07/26-03:13] |
| | Response : Edit : Delete |
|
| hi all, | trying to find something that will enable computer science students to program AIBOs at high levels. | been having difficulty getting pyro to work on mac os x (any help there would be appreciated, too!). my | main question, though, is whether or not it's possible to get the AIBO to recognize the orange ball (like | the one used in robocup) or other objects using R-CODE. if so, how can i go about modifying files to | add to the list of items that the AIBO can search for? | | thanks a lot :) | |
| |- 1806 RE:Is it possible to recognize Orange Balls in R-CODE? AiboPet [2005/07/28-01:06] |
| | | Response : HomePage : Edit : Delete |
|
| | > my | | >main question, though, is whether or not it's possible to get the AIBO to | | recognize the orange ball (like | | >the one used in robocup) or other objects using R-CODE. | | | | What model AIBO? | | For the ERS-2x0/31x you can use RcodePlus which gives you access to all 8 CDT | | color ranges, giving you effectively 5 or 6 distinct color ranges [typically | | deep red, pink, flesh, light blue, pukey yellow/green] | | http://www.aibohack.com/210/colors.htm | | | | For the ERS-7 you have only one color detection (the "Pink_Ball"). It is | | triggered by the deep red or pink color ranges (no distinct). You can replace | | one with orange (triggereing on both orange and pink) or replace both | | (triggering only on orange, pink/red will be ignored) | | | | You can tweek those color ranges with AiboTool, or view the current color | | ranges with ABrowser23 (or ABrowser7 for the ERS-7) | | http://www.aibohack.com/123/aibotool.htm | | This changes the color lookup values in "TR.CT". The file format is relatively | | simple. You can do it directly if you understand the YUV color range format | | (very similar to the OPENR SDK CDT parameters) | | | | > if so, how can i go about modifying files to | | >add to the list of items that the AIBO can search for? | | BTW: For more elaborate detection, using the ERS-7 you can use the AiboMind2 | | visual pattern detection. | | For more elaborate object detection running on the dog itself, you may have to | | use the OpenR SDK. | | |
| | +- 1808 RE:Is it possible to recognize Orange Balls in R-CODE? jcbaillie [2005/08/03-11:46] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi, | | You might want to look at URBI: http://www.urbiforge.com | This might help you ;) | | JC Baillie | | >> my | >>main question, though, is whether or not it's possible to get the AIBO to | >recognize the orange ball (like | >>the one used in robocup) or other objects using R-CODE. | > | >What model AIBO? | >For the ERS-2x0/31x you can use RcodePlus which gives you access to all 8 CDT | >color ranges, giving you effectively 5 or 6 distinct color ranges [typically | >deep red, pink, flesh, light blue, pukey yellow/green] | >http://www.aibohack.com/210/colors.htm | > | >For the ERS-7 you have only one color detection (the "Pink_Ball"). It is | >triggered by the deep red or pink color ranges (no distinct). You can replace | >one with orange (triggereing on both orange and pink) or replace both | >(triggering only on orange, pink/red will be ignored) | > | >You can tweek those color ranges with AiboTool, or view the current color | >ranges with ABrowser23 (or ABrowser7 for the ERS-7) | >http://www.aibohack.com/123/aibotool.htm | >This changes the color lookup values in "TR.CT". The file format is relatively | >simple. You can do it directly if you understand the YUV color range format | >(very similar to the OPENR SDK CDT parameters) | > | >> if so, how can i go about modifying files to | >>add to the list of items that the AIBO can search for? | >BTW: For more elaborate detection, using the ERS-7 you can use the AiboMind2 | >visual pattern detection. | >For more elaborate object detection running on the dog itself, you may have to | >use the OpenR SDK. | |
| +- 1807 RE:Is it possible to recognize Orange Balls in R-CODE? jcbaillie [2005/08/03-11:40] |
| Response : Edit : Delete |
|
>hi all, >trying to find something that will enable computer science students to program AIBOs at high levels. >been having difficulty getting pyro to work on mac os x (any help there would be appreciated, too!). my >main question, though, is whether or not it's possible to get the AIBO to recognize the orange ball (like >the one used in robocup) or other objects using R-CODE. if so, how can i go about modifying files to >add to the list of items that the AIBO can search for? > >thanks a lot :) |
| V 1804 New R-CODE manual in Spanish outsider [2005/07/25-21:00] |
| Response : HomePage : Edit : Delete |
|
Hi everybody. I have created a complete manual for R-CODE and released on my personal web-page. At present it is only available in Spanish, eventhough I have a previous version in English (more limited) also published there. You can find it at www.ouroboros.org Translators to other languages are requested. Hope it helps someone!. |
| V 1801 Please help me to convert the file format holymoon [2005/07/21-08:58] |
| | Response : Email : Edit : Delete |
|
| Is
there a program that can convert MTN format from Master Studio | | into MOT format? | | if do, please tell me the access to the program | | The MOT file is from tekkotsu for edit motion | | But it is not easy to edit | | So i think if there exist a program could convert mtn file to mot file | | Then i could just edit the motion on master studio | | Thanks for help | |
| +- 1802 RE:Please help me to convert the file format ejt [2005/07/21-11:41] |
| | Response : HomePage : Email : Edit : Delete |
|
| Or
alternatively, if anyone has an open-source parser for .mtn files, we
could graft that into | MotionSequence's loader... | |
| +- 1803 RE:Please help me to convert the file format AiboPet [2005/07/22-00:43] |
| Response : HomePage : Edit : Delete |
|
>Or
alternatively, if anyone has an open-source parser for .mtn files,... The "MoNet" sample in the OpenR SDK has a MTN reading library (libMTN) -- the sample reads MTN files from the ODA archive, and plays them directly. My MTN editor (AEditor23) and John of Dogsbody's Skitter has similar MTN file reading and writing code (some guesswork since they were done years ago). I recommend looking at the official SDK. Also has a brief .TXT file on the format. |
| V 1800 A new AIBO Remote Framework FAQ has been added openrsupport@(Administrator) [2005/07/21-03:02] |
| Response : Edit : Delete |
|
A
new item in the AIBO Remote Framework FAQ has been added. Q. Can I control AIBO using AIBO Remote Framework over the Internet? A. You can control your AIBO at home over the Internet by using a wireless LAN router that has a "Remote Access" function. https://openr.aibo.com/openr/eng/perm/rfw/regi_faq3.php4#C0008 Thank you, AIBO SDE Support |
| V 1796 RFW: Sending Image data to wmv file dbroomfield [2005/07/18-23:28] |
| | Response : HomePage : Email : Edit : Delete |
|
| Has
anyone got a sample of how to take an image stream from Remote Framework | and store it as a wmv file (like the AIBO Entertainment Player does) | | Any pointers? I've started looking at the WM9 SDK, is it just a case of | passing the WM_IMAGE_UPDATE to the stream? | |
| +- 1799 RE:RFW: Sending Image data to wmv file AiboPet [2005/07/21-01:35] |
| Response : HomePage : Edit : Delete |
|
>Has
anyone got a sample of how to take an image stream from Remote Framework >and store it as a wmv file (like the AIBO Entertainment Player does) > >Any pointers? I've started looking at the WM9 SDK, is it just a case of >passing the WM_IMAGE_UPDATE to the stream? The data is JPEG format. The easy (but manual) way is to save the results of WM_IMAGETP_UPDATE/ImageGetData as separate .jpg files. Then use one of the many movie editing tools to import a list of JPGs and turn them into a movie. That way you can adjust things like the frame rate, and optionally add sound. For an automatic version, Yes, you can do it with the WM9 SDK. The APIs are overly complicated IMHO, but there are some good samples in the SDK. ---- FWIW: Here's an older sample using the slightly older AVISTREAM APIs and the ERS-2x0 AIBOs. It takes the JPGs, and tacks them on the end of the AVI. http://www.aibohack.com/rcode/index.html#LAN_SAMPLES -- look for the "AiboCam" sample. Needs some adapting to work for the ERS-7 |
| V 1774 Transfer function of the Joints mcmv200i [2005/07/01-22:48] |
| | Response : Email : Edit : Delete |
|
| Hello Everybody! | | I noticed one can set PID parameters of each joint of Aibo. Normally you can | only adjust an controller if you have a mathematical model of the controlled | system. Okay, we cannot build this model because we should not know some | physical parameters of Aibo. But would it maybe be possible to get the transfer | function of the joints to make the adjustment of the PID paramerts easier? | (Maybe some OPEN-R-support member or somebody who has got a good model?) | | Best Regards | Michael | (The article that 2005/07/01-06:31 posted, was modified by Author) | |
| +- 1788 Dear OPEN-R-support... mcmv200i [2005/07/11-21:22] |
| | Response : Email : Edit : Delete |
|
| ...a
short answer like "Sorry, we cannot give you the transfer function
because | of commercial reasons" or "Yes.. here it is: G(s)=..." or "Please wait, we can't | say right now, but might give you later" would be nice! | | Thanks | |
| +- 1790 RE:Dear OPEN-R-support... openrsupport@(Administrator) [2005/07/13-01:47] |
| | Response : Edit : Delete |
|
| I'm
sorry. We cannot give you the transfer functions for commercial reasons. | Please see our FAQ. http://openr.aibo.com/openr/eng/no_perm/faq_openrsdk.php4#A0048 | | Thank you, | | AIBO SDE Support | |
| +- 1795 RE: Dear OPEN-R-support... mcmv200i [2005/07/13-22:29] |
| Response : Email : Edit : Delete |
|
Thank
you! I know this FAQ entry, but yet I asked because this the transfer function is a little more abstract than the hardware details and I thought maybe you could disclose this. But of course, I can understand it from Sony's point of view. Thanks for your answer! |
| V 1794 OPEN-R SDK with DEV-C++ or Eclipse randeep [2005/07/13-16:13] |
| Response : Edit : Delete |
|
Hi All, Has anyone configured Cygwin, MIPS, and OPEN-R tools to work in DEV-C++ or Eclipse IDE. Please let me know. Thanks. |
| V 1756 Remote Processing OPEN-R SDK randeep [2005/06/17-18:58] |
| | Response : Edit : Delete |
|
| Hi, | | I plan to work on OPEN-R SDK and want to use Remote Processing thats only | available on linux. | | Which linux distribution you recommend (RedHat/ Fedora, Mandrake, SuSe etc.) on | which OPEN-R and other libs compile easily. | | Please let me know. | | Thanks, | |
| +- 1757 RE:Remote Processing OPEN-R SDK AiboPet [2005/06/18-00:21] |
| | Response : HomePage : Edit : Delete |
|
| FWIW:
I prefer Windows first, RedHat second. | | >I plan to work on OPEN-R SDK and want to use Remote Processing thats only | >available on linux. | "Remote Processing OpenR" also runs under cygwin (ie. under Windows) along | with all the build tools. | https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq2.php4#A0028 | | If you have other code that requires Linux, then pick the distribution based | on the needs of that package. | |
| +- 1763 RE:Remote Processing OPEN-R SDK randeep [2005/06/24-17:56] |
| | Response : Edit : Delete |
|
| I
could setup (./setup-rp-openr) Remote Processing OPEN-R under SuSe
Linux and | in Windows using Cygwin. When I start (./start-rp-openr) I get the following: | | On Linux: | | linux:/usr/local/OPEN_R_SDK/RP_OPEN_R/bin # ./start-rp-openr | [pid:7757,msqid:98307,oid:0x00018003] oserviceManager | [pid:7758,msqid:131076,oid:0x00020004] tcpGateway | MS/OPEN-R/MW/CONF/OBJECT.CFG: No such file or directory | | On Windows: | | $ ./start-rp-openr | msgget: Function not implemented | | I dont know why the responses on the two platforms are different? Is there a | sample program which shows how to use Remote Processing OPEN-R. | | Thanks | |
| +- 1793 RE:Remote Processing OPEN-R SDK openrsupport@(Administrator) [2005/07/13-01:54] |
| Response : Edit : Delete |
|
Yes,
there are sample programs which use Remote Processing OPEN-R. Please see chapter 4 in the Installation Guide. It will show you how to use Remote Processing OPEN-R with the ObjectComm and LMasterRSlave sample programs. There is also a description of Remote Processing in the Programmers Guide (chapter 6). Thank you, AIBO SDE Support |
| V 1779 R-CODE ARRAYS? matan325 [2005/07/08-07:28] |
| | Response : Email : Edit : Delete |
|
| Hello! | | Is there any plans whatsoever on implementing Array support in R-CODE (for | ERS-7)? Obviously this would make R-CODE (_a lot_ more) useful for all kinds of | things. It feels kind of unnessescary to have to go trough OPEN-R just because | of this, even if I only need to implement some really basic things. (I am | currently designing some laborations for an undergraduate university AI course, | and I want to keep things simple.) | | I was very glad when I noticed there was something called "R-CODE PLUS" | (unofficial) which included support for Arrays, and very disappointed when I | realized that it did not exist for ERS-7, but only for the earlier robot models. | The author of R-CODE PLUS said there is no plans on releasing this for the ERS-7 | though. | | Have a nice day! | | Mats Andren, Sweden | (The article that 2005/07/08-07:27 posted, was modified by Author) | |
| +- 1791 RE:R-CODE ARRAYS? openrsupport@(Administrator) [2005/07/13-01:52] |
| Response : Edit : Delete |
|
Thank
you for showing your interest. Unfortunately, there isn't a current plan to add array support into R-CODE. We will inform you on this BBS if and when a new release of R-CODE is available. R-CODE is kind of like assembly. Higher level languages may be able to create R-CODE (like a compiler). Thank you, AIBO SDE Support |
| V 1785 R-CODE NoFallDown sopl [2005/07/11-04:34] |
| | Response : HomePage : Edit : Delete |
|
| I
have a problem with the motion 1291 (karate3) with the AIBO 210. After
the | motion he think it tilted and he stand up. With RCodePlus I can switch the | problem off with the instruction "SET:NoFallDown:1". As with R-CODE SDK goes? | |
| +- 1789 RE:R-CODE NoFallDown AiboPet [2005/07/12-01:10] |
| Response : HomePage : Edit : Delete |
|
"SET
NoFallDown 1" is supported in *all* the current versions of RCODE/RCodePlus for the 2x0/31x and ERS-7 [an undocumented Sony feature, not a "Plus" feature] For the ERS-7 version, be sure to get the most recent download. General warning: turning off fall down protection ("SET NoFallDown 1") and then re-enabling it later ("SET NoFallDown 0") can sometimes confuse the robot. |
| V 1786 I still have problem with MEditor, and no clues how o fix it tevye [2005/07/11-17:29] |
| | Response : Email : Edit : Delete |
|
| I
don't know if i' being ignored or no one has an answer to my problem,
but a | post just saying something would help, even if it is just to say "not a clue". | | Regards | |
| +- 1787 RE:I still have problem with MEditor, and no clues how o fix mcmv200i [2005/07/11-21:18] |
| Response : Email : Edit : Delete |
|
Hello, connecting Medit with Aibo via Wavelan is not availible at the moment. See the Medit manual on page 9: >> The contents control dialog is normally used for interactive testing with an AIBO connected via Wireless LAN. Unfortunately, this function is not active in the current version of MEdit, so we will not cover it here. Please disregard this window in your daily use. << |
| V 1782 Wireless lan card for a 210A alevy222 [2005/07/09-12:38] |
| | Response : Email : Edit : Delete |
|
| Hi
is there a wireless lan card other than the sony Aibo ERS-201D1 that
will | work in a 210A or is the only one the sony one | | Thanks | | Al | |
| +- 1784 RE:Wireless lan card for a 210A AiboPet [2005/07/11-02:57] |
| Response : HomePage : Edit : Delete |
|
>Hi
is there a wireless lan card other than the sony Aibo ERS-201D1 that
will >work in a 210A or is the only one the sony one The ERS-201D1 or the Sony VAIO PCWA-C150S (almost exactly the same) Both are discontinued and getting hard to find. Good luck More recent VAIO cards (now supporting 802.11g) do not work. Cheaper card FAQ: http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi? ubb=get_topic;f=10;t=000020 |
| V 1781 AiboCam project - Fundable listing online PseudoThink [2005/07/09-06:09] |
| | Response : HomePage : Email : Edit : Delete |
|
| Check
it out! Add a comment, or even better, a contribution pledge! Only two | more contributions are needed for it to become a publically listed fundraiser! | | http://www.fundable.org/groupactions/aibocam | |
| +- 1783 RE:AiboCam project - Fundable listing online PseudoThink [2005/07/10-15:07] |
| Response : HomePage : Email : Edit : Delete |
|
>Check
it out! Add a comment, or even better, a contribution pledge! Only two >more contributions are needed for it to become a publically listed fundraiser! > >http://www.fundable.org/groupactions/aibocam Is anyone even interested in this idea? The potential to use Aibo as a remotely-controllable mobile webcam seemed so cool that it was the entire reason I bought one in the first place. The software component (web pages, server component, Aibo control stick image, etc.) was free and a hardware kit was available for ~$100 that would allow you modify the Aibo to work on four powered wheels, would you buy it? |
| V 1780 The Academic's Corner on the European AIBO website openrsupport@(Administrator) [2005/07/09-01:47] |
| Response : Edit : Delete |
|
The
Academic's Corner is available on the European AIBO website. You will
find a video of AIBO in action and discover how three leading universities use AIBO for the action-packed RoboCup competitions and for research. Discover how AIBO inspires students and researchers. Click here http://www.eu.aibo.com/5_1_casestudies.asp?language=en Thank you, AIBO SDE Support |
| V 1761 wireless communication ventresca7 [2005/06/22-17:22] |
| | Response : Email : Edit : Delete |
|
| Hi
does anyone now the size of a UDP package that I can send from AIBO? i
am | trying to send data over the wifi to a java server application running on a | windows xp machine, but i am able to receive only 400 bytes per package. Is | this something i can configure on the aibo udp endpoint source files or is it | a problem with the java server?? | | Thanks a lot | |
| +- 1762 RE:wireless communication AiboPet [2005/06/23-23:56] |
| | Response : HomePage : Edit : Delete |
|
| > ... or is it | >a problem with the java server?? | | That would be my guess. | | I use UDP transfers for JPG image data all the time (10 -> 30 KB typically). | If you are sending it as one chunk from AIBO in a "UDPEndpointSendMsg" it | *should* get to your PC in one piece (eg: a standard Winsock "recvfrom") | If you are receiving UDP packets in some other way -- I don't know. | |
| +- 1778 RE:wireless communication ventresca7 [2005/07/07-22:03] |
| Response : Email : Edit : Delete |
|
Hi
I am trying to send data from Aibo to a udp server application running
on a PC, and i need to use a port between 9400 and 9500. While port 9876 works fine, in cannot send data to port 9476. Is there some reason why this port cannot be used on Aibo? and if it is reserved, which other port between 9400 and 9500 can i use? thanks andrea >> ... or is it >>a problem with the java server?? > >That would be my guess. > >I use UDP transfers for JPG image data all the time (10 -> 30 KB typically). >If you are sending it as one chunk from AIBO in a "UDPEndpointSendMsg" it >*should* get to your PC in one piece (eg: a standard Winsock "recvfrom") >If you are receiving UDP packets in some other way -- I don't know. |
| V 1776 Looking for ERS7M2 or 7 w/upgrade nighthawk45750 [2005/07/04-11:55] |
| | Response : HomePage : Email : Edit : Delete |
|
| I am
seeking an Aibo ERS7M2 or 7 with a mind 2 upgrade. I usually like to buy | new, but 2000 dollors is very steep for me. If you have any info on a gently | used Aibo, please email or post. Thank you. | |
| +- 1777 RE:Looking for ERS7M2 or 7 w/upgrade openrsupport@(Administrator) [2005/07/05-15:10] |
| Response : Edit : Delete |
|
Thank
you for your interest in AIBO. However, this bulletin board is strictly for discussion about AIBO software development. Please post your question at an appropriate 3rd party site. AIBO SDE Support |
| V 1768 SEMANTICS RFW jcol1817 [2005/06/29-23:39] |
| | Response : Email : Edit : Delete |
|
| Hello, | | I have one problem with semantics in RFW. I need to get the position of the | ball, and I use input semantics messages... but the problem is that when I get | subinfo[0] and subinfo[1] (the horizontal and vertical degrees of the ball) I | always get the same position... anyone can help me? | | Thanks | |
| +- 1770 RE:SEMANTICS RFW AiboPet [2005/06/30-00:00] |
| | Response : HomePage : Edit : Delete |
|
| >>
I need to get the position of the | >ball, and I use input semantics messages. | | Are you using SEMID_PINK (no position) or SEMID_BALL (should have position) ? | |
| +- 1773 RE:SEMANTICS RFW jcol1817 [2005/07/01-03:31] |
| | Response : Edit : Delete |
|
| I'm
using SEMID_BALL... and I have always been getting in the same info...
it | not deppends on the ball`s position... (h=10,v=0,dist=25) | | I have thought I can use some kind of image processor to get the ball's | position, I think I could use some C++ API's or something like this... | Do you know where I could find something like this? | |
| +- 1775 RE:SEMANTICS RFW openrsupport@(Administrator) [2005/07/02-06:27] |
| Response : Edit : Delete |
|
We
apologize for your problems with SEMID_XXX. SEMID_XXX returns the horizontal/vertical position and estimated distance (cm). These parameters use "Robot Coordinates". The center of "Robot Coordinates" is on the floor just under the center of AIBO's body. Zero angle is in front of the body (not head). So, if the pink ball is 0cm from the camera, the distance will be the length from the center of the body to AIBO's nose. (But AIBO cannot recognize it as a ball if it's 0cm away) If AIBO's head doesn't move, the ball position may not change. Because the ball can move a lot, it may be out of the camera view. If the ball moves far from camera, AIBO cannot recognize it as a ball. The recommended usages are 1. Target recognition ON. 2. Search & track the pink ball. 3. (If you want to approach the ball) Use CVAIBO::DoApproach(). Thank you, AIBO SDE Support |
| V 1772 Motion Editor doesn't connecto to ers7 tevye [2005/06/30-20:30] |
| Response : Email : Edit : Delete |
|
Everytime
i try to connect by wireless to my Aibo M2 SP3 using Medit i always get the same error: "OnLookupNetService Error" i have the wireless working ok with AEP. Am i missing a step? Regards |
| V 1766 R-Code NEWBIE swei80 [2005/06/29-13:48] |
| | Response : Email : Edit : Delete |
|
| Hi,
I am currently a student understaking a thesis. I am supposed to
implement | BDI concept to enable autonomous behavior in the ERS-7 using R-CODE. | | Is there anyone who have done thing before or think about this before? I will | be using the camera as visual input, microphone as voice input and touch | sensor to gather info of the environment. I am thinking how to use the info to | generate behavior. | | Thank you for any help~ | |
| +- 1769 RE:R-Code NEWBIE AiboPet [2005/06/29-23:58] |
| Response : HomePage : Edit : Delete |
|
>Hi,
I am currently a student understaking a thesis. I am supposed to
implement >BDI concept to enable autonomous behavior in the ERS-7 using R-CODE. There are is no way to add advanced/custom inputs to RCODE, so you are limited to what is built-in. First check the ERS-7 RCODE reference (part of the RCODE download) to make sure you can get data input to do the job. Otherwise you may need another tool (like the OpenR SDK or RFW) > I will >be using the camera as visual input RCODE supports finding/tracking the pink ball and/or pink bone [and the limited Face recognition] >, microphone as voice input RCODE supports ~50 voice commands. http://www.aibohack.com/rcode/rc_vcmd.htm It also can handle some special tones/tunes. > and touch >sensor to gather info of the environment. The touch sensors and IR distance sensors (obstacle detection) are well supported. Programming in RCODE can be very easy, but also can be a big pain. For very simple programs, YART makes things very easy http://www.aibohack.com/ers7/yart7.htm For more advanced RCODE scripts, you should use a WiFi connection to "telnet" to your AIBO to see how the program is executing (eg: with PRINT statements) |
| V 1767 how to read the position of aibo in the world field wisdom [2005/06/29-22:02] |
| Response : Email : Edit : Delete |
|
when
the aibo was placed on the Robocup field,how can I read its position on the world field? any help will be appreciated. |
| V 1764 Aibo Walk ccrews [2005/06/25-08:06] |
| | Response : Email : Edit : Delete |
|
| Good Evening/Afternoon/Morning, | I am a student at the Air Force Institute of Technology and am using the AIBO | ERS-7 for my thesis work. My first step is to program the AIBO to walk and | run its camera. I've looked over the Tekkotsu code, but don't want to work | using the GUI or the Start Behavior classes. I just need basic walking loop. | Is there any documentation to tell me how to program it using just the OPENR | platform? Can I extract code from the WalkMC classes in Tekkotsu? Any input | of the best way to approach would be greatly appreciated, since the bulk of my | work is in other areas (real time image processing). | | Thanks | Carrie | |
| +- 1765 RE:Aibo Walk AiboPet [2005/06/27-03:22] |
| Response : HomePage : Edit : Delete |
|
>....
I just need basic walking loop. >Is there any documentation to tell me how to program it using just the OPENR >platform? Can I extract code from the WalkMC classes in Tekkotsu? The core walk engine used by Tekkotsu (and others) is the CMU Robocup winning walk algorithm (CMPack 02) Tekkotsu wrapped it to work within their framework. If you don't plan on using Tekkotsu, I suggest using the CMPack OpenR modules directly. http://www-2.cs.cmu.edu/%7Erobosoccer/legged/ (now CMPack 04) That's what I did for Walkbench and AiboSpeedRacer (straight Open-R SDK programs, using the CMPack '02 walk engine stripped down to the basics) http://www.aibohack.com/openr_sdk/walkbench.htm http://www.aibohack.com/openr_sdk/speedracer.htm http://www.aibohack.com/openr_sdk/SpeedRacer01_SRC.zip BTW: I would love it if Sony would release a "SoccerLion" sample for the ERS-7 so people would not have to reinvent this all the time - which is limited to stressful "knee" walking. An option for ERS-210/220 owners. |
| V 1568 Network Problems birman1 [2004/12/31-23:33] |
| | Response : Edit : Delete |
|
| Can
anyone help with information on known aibo / router problems and their | resolution. | | I have been trying to network my ERS-7 "Monty" for some time. I have a | Speedtouch 510 ADSL wireless router. The router has found "Monty" and put his | mac address in the allowed access table but does not allocate an IP address. | Even setting him up with a static IP address still does not help, ping to the | address gets no response. I have seen some sugestions that an SSID with mixed | case might be a problem - but this seems a bit peverse. | | Anyone got any ideas ? | |
| +- 1571 RE:Network Problems AiboPet [2005/01/03-02:19] |
| | Response : HomePage : Edit : Delete |
|
| >Can
anyone help with information on known aibo / router problems and their | >resolution. | | Check the Aibo-Life.org forums, in particular the WiFi LAN FAQs | http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi?ubb=forum;f=10 | | ----- | | >I have been trying to network my ERS-7 "Monty" for some time. I have a | >Speedtouch 510 ADSL wireless router. The router has found "Monty" and put his | >mac address in the allowed access table but does not allocate an IP address. | | In general I recommend people start with the most basic configuration (the | default for ERS-210) with standard SSID/WEP, hard coded IP address (no DHCP). | Get ping working, then http://10.0.1.100 web access. | | After it is working, change things one at a time as you see fit. | Adding support for AiboMail can be much more difficult. | | WLANDFLT.TXT=> | ---- | HOSTNAME=AIBO | ETHER_IP=10.0.1.100 | ETHER_NETMASK=255.255.255.0 | IP_GATEWAY=10.0.1.1 | ESSID=AIBONET | WEPENABLE=1 | WEPKEY=AIBO2 | APMODE=2 | CHANNEL=3 | ----- | |
| +- 1573 RE:Network Problems birman1 [2005/01/04-02:10] |
| | Response : Edit : Delete |
|
| snip | >------------------------------------------------- | > | >In general I recommend people start with the most basic config | >---- | >HOSTNAME=AIBO | >ETHER_IP=10.0.1.100 | >ETHER_NETMASK=255.255.255.0 | >IP_GATEWAY=10.0.1.1 | >ESSID=AIBONET | >WEPENABLE=1 | >WEPKEY=AIBO2 | >APMODE=2 | >CHANNEL=3 | >----- | Thanks for taking the trouble to reply. It is very difficult to start at this | very basic level as there are other computers using the network, so I would | need a very good justification to change the ESSID or wep key. | | Several things strike me from the above setup: | | My setup does not specify a channel number, the router sellects the best | channel on connection. Can Aibo follow this regime or does he have to use a | fixed channel? | | I use a hex wepkey of the form WEPKEY=0x0000000000 in my WLANCONF | | The fact that the router has found the Aibo mac address suggests to me that | the aibo name and wepkey are not causing problems. | | Having failed with DHCP I gave the dog a fixed IP address in the DNS table. | Ping fails but the router seems to admit the address is a possibility and | tries 3 packets | | Any ideas - I am probably doing something daft ........ I do wonder about | chipset inconpatiblities and the like. The only references I have found to | aibo and speedtouch510 are Dutch, but as far as I can work out contact was | made! | |
| +- 1574 RE:Network Problems AiboPet [2005/01/04-02:55] |
| | Response : HomePage : Edit : Delete |
|
| >
It is very difficult to start at this | >very basic level as there are other computers using the network | | If possible, setup an ad-hoc connection first -- using a separate WiFi card | [borrow one until it is working] | | Unless your existing WiFi setup uses up many separate WiFi channels, there | should be enough room for them to co-exist. | | Then you can selectively move over a working setup from the ad-hoc to your | firewalled router (one piece at a time, until you isolate the specific router | setting) - or decide to keep it as a separate net. | | [better than guessing IMHO] | | ----- | >so I would | >need a very good justification to change the ESSID or wep key. | [somewhat of a 'catch 22'] | | If you are in an organization that polices WiFi that much, you probably | shouldn't be using AIBO on it ;-> | (possible security and bandwidth hog - it is only 802.11b remember) | | ---- | BTW: It usually will find the channel from the access point, but for the ad- | hoc scenario I find it helps to hard-code it. | |
| +- 1577 RE:Network Problems birman1 [2005/01/05-05:36] |
| | Response : Edit : Delete |
|
| Thanks
for the useful suggestion of an ad-hoc setup. I have a spare notebook | PC with a WiFi card ready to put in. Will probably find the time to try this | next weekend and will report on progress. | > | >If you are in an organization that polices WiFi that much, you probably | >shouldn't be using AIBO on it ;-> | | Its actualy my home network ....... but I need it to be up and running. | | >>BTW: It usually will find the channel from the access point, but for the ad- | >hoc scenario I find it helps to hard-code it. | | Noted | |
| +- 1760 RE:Network Problems ventresca7 [2005/06/22-17:21] |
| Response : Email : Edit : Delete |
|
Hi
does anyone now the size of a UDP package that I can send from AIBO? i
am trying to send data over the wifi to a java server application running on a windows xp machine, but i am able to receive only 400 bytes per package. Is this something i can configure on the aibo udp endpoint source files or is it a problem with the java server?? Thanks a lot |
| V 1728 ERS7 Flash / Reset Sauber [2005/05/23-13:41] |
| | Response : Edit : Delete |
|
| Hi, | | Is there a way that you can reset or flash an ERS-7 to its original state? We | have a problem where the openr primitives won't initialise, we were not doing | anything before this occurred (i.e. all the code was doing is counting vision | frames!). Then the next time we booted it generated primitive errors, which | results in the robot shutting down. My gut feeling is that a complete reset of | the internal software maybe able to fix this, is this possible ? | | Michael | |
| +- 1729 RE:ERS7 Flash / Reset AiboPet [2005/05/23-23:29] |
| | Response : HomePage : Edit : Delete |
|
| >Is
there a way that you can reset or flash an ERS-7 to its original state? | The bulk of the ERS-7 firmware is not updatable. | | There is a small area of internal Flash ROM that stores AiboMind personality | read-write data (so the core part of the personality remains even if you | switch memory sticks). This should not influence the behavior when running | RCODE or OpenR SDK programs. | | ---- | > We | >have a problem where the openr primitives won't initialise, we were not doing | >anything before this occurred... | | They can be other problems. | The most common is the battery, followed by broken or flakey connections to | the joints [more common on the 2x0 which has removable joints] | | My advice: charge up the battery, take out any memory stick and boot up the | dog in minimal clinic/diagnostic mode. If that doesn't work - and you want to | try to fix it yourself - see which primitives are not initializing (ie. this | may help you determine which cable is having problems) | Good luck | |
| |- 1731 RE:ERS7 Flash / Reset Sauber [2005/05/24-10:54] |
| | | Response : Edit : Delete |
|
| | Hi, | | I have with a varitey of memory sticks (including none) and batteries. It | | pretty dies on the first primitive that is contacted, I can get back to you | | exactly when it has the problem, but that doesnt really make that much | | difference. I suppose my issue is that the robot did not move and inch from the | | boot that work and the boot that failed. I understand that if the cable was | | loose that it might have taken nothing to break it but I really think in this | | case that is not the problem (i.e. this robot is quite new and has barely been | | used). | | It was suggested that it could be due to the use of non-sony memory stick | | writer (or possibly the memory stick was ejected before the shutdown complete). | | Are either of these capable of generating this problem? If so we can make sure | | that it never happens again. | | | | Micahel | | | | >>Is there a way that you can reset or flash an ERS-7 to its original state? | | >The bulk of the ERS-7 firmware is not updatable. | | > | | >There is a small area of internal Flash ROM that stores AiboMind personality | | >read-write data (so the core part of the personality remains even if you | | >switch memory sticks). This should not influence the behavior when running | | >RCODE or OpenR SDK programs. | | > | | >---- | | >> We | | >>have a problem where the openr primitives won't initialise, we were not doing | | >>anything before this occurred... | | > | | >They can be other problems. | | >The most common is the battery, followed by broken or flakey connections to | | >the joints [more common on the 2x0 which has removable joints] | | > | | >My advice: charge up the battery, take out any memory stick and boot up the | | >dog in minimal clinic/diagnostic mode. If that doesn't work - and you want to | | >try to fix it yourself - see which primitives are not initializing (ie. this | | >may help you determine which cable is having problems) | | >Good luck | | |
| | +- 1740 RE:ERS7 Flash / Reset AiboPet [2005/05/27-00:50] |
| | | Response : HomePage : Edit : Delete |
|
| | >
It was suggested that it could be due to the use of non-sony memory
stick | | >writer (or possibly the memory stick was ejected before the shutdown | | complete). | | Shouldn't be a problem. | | [there was a flash storage problem with the ERS-11x - but only if you ran | | certain software] | | | | To clarify: using a non-Sony stick reader or ejecting the stick when in use | | can (and quite often will) corrupt the stick data, perhaps making *the memory | | stick* unusable. | | | | However it shouldn't have any lasting effects on the dog. In all these cases, | | test without a stick in the dog. If it will boot and stretch (and do the basic | | builtin diagnostics) then the memory stick is most likely the problem. It may | | be a stick format or copy protection problem (if it stops before any of your | | software runs), or it may be the custom software. | | |
| | +- 1742 RE:ERS7 Flash / Reset Sauber [2005/05/27-15:18] |
| | Response : Edit : Delete |
|
| Hi, | It does the same thing with no memory stick, so its a problem with the robot. | | Michael | >> It was suggested that it could be due to the use of non-sony memory stick | >>writer (or possibly the memory stick was ejected before the shutdown | >complete). | >Shouldn't be a problem. | >[there was a flash storage problem with the ERS-11x - but only if you ran | >certain software] | > | >To clarify: using a non-Sony stick reader or ejecting the stick when in use | >can (and quite often will) corrupt the stick data, perhaps making *the memory | >stick* unusable. | > | >However it shouldn't have any lasting effects on the dog. In all these cases, | >test without a stick in the dog. If it will boot and stretch (and do the basic | >builtin diagnostics) then the memory stick is most likely the problem. It may | >be a stick format or copy protection problem (if it stops before any of your | >software runs), or it may be the custom software. | |
| +- 1759 RE:ERS7 Flash / Reset ventresca7 [2005/06/22-17:11] |
| Response : Email : Edit : Delete |
|
HI
is there a way of freeing that "small area of internal Flash ROM that stores AiboMind personality read-write data" that you mentioned?? I save jpeg frames in the mw\data\p directory on the stick, but these never refresh and even if i reboot the robot it catches the old jpeg frame without using the present ofbk. In some way it is similar to what sony's ImageObserver sample does... what can be the reason for this? thanks andrea >>Is there a way that you can reset or flash an ERS-7 to its original state? >The bulk of the ERS-7 firmware is not updatable. > >There is a small area of internal Flash ROM that stores AiboMind personality >read-write data (so the core part of the personality remains even if you >switch memory sticks). This should not influence the behavior when running >RCODE or OpenR SDK programs. > >---- >> We >>have a problem where the openr primitives won't initialise, we were not doing >>anything before this occurred... > >They can be other problems. >The most common is the battery, followed by broken or flakey connections to >the joints [more common on the 2x0 which has removable joints] > >My advice: charge up the battery, take out any memory stick and boot up the >dog in minimal clinic/diagnostic mode. If that doesn't work - and you want to >try to fix it yourself - see which primitives are not initializing (ie. this >may help you determine which cable is having problems) >Good luck |
| V 1755 Question about SetData caguero [2005/06/09-04:41] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I have a question about how SetData function works for communicate various | Open-R objects. According to the manual, SetData copy the data to a shared | buffer and then invoking NotifyObservers, the Open-R objects are advised and | they can access data. | | My question is what happens if an object invokes SetData (with data1) and | notifyObservers, and next (the next instructions) make the same operation | invoking SetData with data2 and NotifyObservers. | | The result depends of how quickly takes data in the other side of communication | or contrary the second copy made for SendData don't overwrite the first? | | Thanks in advance | |
| +- 1758 RE:Question about SetData AiboPet [2005/06/18-00:33] |
| Response : HomePage : Edit : Delete |
|
Potentially messy. The "ready" state of the observer matters as well. If the observer is ready when the NotifyObservers is called, the observer's handler is called right away. Otherwise the data is saved to wait for the observer to be ready. If the SetData has limited lifetime information (eg: other data in shared memory), you may have other problems. When you have a number of observers it can get more complicated. Suggestions: #1) if you are writing both the subject and observer, make the data structure more general so you can do it with a single SetData/NotifyObservers. Make all the data self-contained (so it can be saved or dropped without harm) #2) if you need two separate SetData/NotifyObservers, then check the ready status ("IsReady()" or "IsAllReady()") before the SetData/NotifyObservers. If they are all ready in both cases (IsAllReady) you won't have any congestion and no chance of lost data. #3) (don't know if this will be needed, but) you can tweek the task priority of any of the app modules to work-around any potential deadlock situations - usually all the modules have the same priority |
| V 1506 Failed to port ERS-210 program to ERS-7 wangy22 [2004/11/15-12:08] |
| | Response : HomePage : Email : Edit : Delete |
|
| I've
written a program to control AIBO ERS-210 remotely via wireless network, | during last summer. It contains two parts, which are the server to be executed | on AIBO and the client to be run on Linux at local pc. | | Current I am porting the server part to AIBO ERS-7. The main problem is the | ERS-7, which is running server part, cannot move its legs at all. I know it is | really tough and bad coding, as I am not a good programmer and I am really busy | with my university study. | | So is anyone who is interested in looking at my code and help me to solve the | various problems? | | The sourcecodes can be downloaded at: | | http://msw.mcmaster.ca/~wangy/aibo/ | | The required environment is: | | 1. Latest OPEN-R SDK (Of course) | 2. Any Linux | 3. SDL-1.2.6+ for video streaming (Can be downloaded at www.libsdl.org) | 4. eSound 0.2+ for audio streaming (optional) | | I don't remember what other libraries are required, but should be easy to find out. | | The server for ERS-210 is working on my ERS-210 AIBO, I haven't test it with | other AIBOs. So test them at YOUR OWN RISK....sorry about that.... | | Any help would be appreciated! | | Sincere, | | Yu Wang | (The article that 2004/11/15-12:02 posted, was modified by Author) | |
| +- 1750 RE:Failed to port ERS-210 program to ERS-7 ventresca7 [2005/06/06-20:06] |
| | Response : Email : Edit : Delete |
|
| Hi
i am a researcher at niversity of rome and would like to port liveo
video | streaming (at least 10 frames per second) on the OPENR AIBO architecture i am | using. I saw from your message that you used this program: | | 3. SDL-1.2.6+ for video streaming (Can be downloaded at www.libsdl.org) | | is it compatible with sony aperios? in order to only implement a video client | on aibo and a server on the PC to send nice mpeg like aibo mind2? | | thanks a lot | | andrea | | >I've written a program to control AIBO ERS-210 remotely via wireless network, | >during last summer. It contains two parts, which are the server to be executed | >on AIBO and the client to be run on Linux at local pc. | > | >Current I am porting the server part to AIBO ERS-7. The main problem is the | >ERS-7, which is running server part, cannot move its legs at all. I know it is | >really tough and bad coding, as I am not a good programmer and I am really | busy | >with my university study. | > | >So is anyone who is interested in looking at my code and help me to solve the | >various problems? | > | >The sourcecodes can be downloaded at: | > | >http://msw.mcmaster.ca/~wangy/aibo/ | > | >The required environment is: | > | >1. Latest OPEN-R SDK (Of course) | >2. Any Linux | >3. SDL-1.2.6+ for video streaming (Can be downloaded at www.libsdl.org) | >4. eSound 0.2+ for audio streaming (optional) | > | >I don't remember what other libraries are required, but should be easy to | find out. | > | >The server for ERS-210 is working on my ERS-210 AIBO, I haven't test it with | >other AIBOs. So test them at YOUR OWN RISK....sorry about that.... | > | >Any help would be appreciated! | > | >Sincere, | > | >Yu Wang | >(The article that 2004/11/15-12:02 posted, was modified by Author) | |
| +- 1754 RE:Failed to port ERS-210 program to ERS-7 AiboPet [2005/06/09-00:18] |
| Response : HomePage : Edit : Delete |
|
>3.
SDL-1.2.6+ for video streaming (Can be downloaded at www.libsdl.org) That's running on the PC. SDL is apparently used by this PC program to take the raw uncompressed YUV images from the AIBO (through WiFi) and draw it on the PC screen. [This can easily be done on your own without relying on "SDL"] ---- >is it compatible with sony aperios? No - SDL is not running on the AIBO. > in order to only implement a video client >on aibo and a server on the PC to send nice mpeg like aibo mind2? See other reply regarding video. Compressed video (JPEG compression) is recommended if you want a good frame rate and full sized images (416x320) |
| V 1383 video encoder laurenshobert [2004/09/15-22:50] |
| | Response : Edit : Delete |
|
| Is
there a motion-jpeg or mpeg encoder available for the open-r framework? | (The article that 2004/09/15-22:48 posted, was modified by Author) | |
| +- 1384 RE:video encoder ejt [2004/09/16-03:27] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Is
there a motion-jpeg or mpeg encoder available for the open-r framework? | >(The article that 2004/09/15-22:48 posted, was modified by Author) | The best that I can suggest is to use one of the video streaming packages, and | then record the frames to disk and encode them on the PC. | I believe there was a robosoccer group at UPenn who did actually implement a | motion-jpeg format, but I'm not sure if that's public. I'm pretty sure AiboPet | has some binaries available to record video to files, if not a video. And you | could always use our Tekkotsu package to record the frames as well. | |
| +- 1387 RE:video encoder AiboPet [2004/09/16-23:52] |
| | Response : HomePage : Edit : Delete |
|
| >The
best that I can suggest is to use one of the video streaming packages,
and | >then record the frames to disk and encode them on the PC. | | I would agree. Doing some simple compression on AIBO itself to reduce the WiFi | bandwidth requirements, but then have a PC do the heavy duty compression | (allowing for a number of formats, easy to do with Windows, see my AiboCam | program if interested) | | You may need to experiment with frame rate and image size (a full color image | data at 25fps is a lot of data) | |
| +- 1751 RE:video encoder ventresca7 [2005/06/06-21:09] |
| | Response : Email : Edit : Delete |
|
| Hi
I am also interested in LIVE video streaming at high frame rate (at
least | 10-15 fps). Where can i get these ready-made video streaming packages? is it | possible to port them in MY open-r code instead of aibomind? | For instance, is it possible to simply add an open-r object to the memory | stick to compress and send the video to the UDP server on the PC just like | aibomind2 does? Is there an open-r object or some source code available for | this from Sony? | | thanks a lot | | >>The best that I can suggest is to use one of the video streaming packages, | and | >>then record the frames to disk and encode them on the PC. | > | >I would agree. Doing some simple compression on AIBO itself to reduce the WiFi | >bandwidth requirements, but then have a PC do the heavy duty compression | >(allowing for a number of formats, easy to do with Windows, see my AiboCam | >program if interested) | > | >You may need to experiment with frame rate and image size (a full color image | >data at 25fps is a lot of data) | |
| +- 1753 RE:video encoder AiboPet [2005/06/09-00:15] |
| Response : HomePage : Edit : Delete |
|
>Hi
I am also interested in LIVE video streaming at high frame rate (at
least >10-15 fps). Where can i get these ready-made video streaming packages? To clarify, from AIBO's perspective "video streaming" is nothing more than sending one image after another from the camera. The capture programs I've seen use either: * no compression YUV - 416x320x3 bytes = 400KB per frame * JPG compression - usually around 20KB per frame (depending on quality), slow to compress * my YUV10 codec - fast and sleazy under 30KB per frame How the image data is sent back to the PC impacts the frame rate too. Of the options, TCP/IP or UDP/IP, UDP is generally faster. That's for the AIBO part. ------ I believe the mention of "video streaming packages" was for the PC part, to get the raw image data (YUV or JPG) and turn it into something else. ======== >possible to port them in MY open-r code instead of aibomind? Yes. Using the Open-R SDK is the only option for your own video processing [AiboMind2/RFW has built in image capture using JPG compression] >For instance, is it possible to simply add an open-r object to the memory >stick to compress and send the video to the UDP server on the PC just like >aibomind2 does? Yes, but you will be doing it much more directly than AiboMind2/RFW. > Is there an open-r object or some source code available for >this from Sony? Not in the Sony Open-R SDK samples (there is a still image capture sample) There are several JPG compression versions and several uncompressed versions for "streaming" Mine is used in AiboSpeedRacer http://www.aibohack.com/openr_sdk/speedracer.htm Source: http://www.aibohack.com/openr_sdk/SpeedRacer01_SRC.zip (direct OpenR SDK JPG+UDP VisionU module and tweeked CMU Walk engine) If using Tekkotsu (not direct OpenR SDK), they have similar features built in and in their samples. |
| V 1743 Member-Function atomic in OPEN-R? mcmv200i [2005/06/02-04:57] |
| | Response : Email : Edit : Delete |
|
| The
Paper "Programming Model Based on Concurrent Objects for the AIBO
Robot" by | Francisco Martin Rico, Rafaela Gonzalez-Careaga, Jose Maria Canas Plaza, and | Vicente Matellan Olivera says that a member-function will not be interrupted by | Aperios. And it says this is archieved by interrupt masking. Is that (still) right? | |
| +- 1745 RE:Member-Function atomic in OPEN-R? AiboPet [2005/06/04-00:22] |
| | Response : HomePage : Edit : Delete |
|
| >...
says that a member-function will not be interrupted by Aperios. | True, from the perspective of that one callback function (typically a member | function in your class) within the one current task / module. | | Within your one module - there is only one execution path (one thread per | task/module) | | void MyClass::DoNotifyEvent1(const ONotifyEvent& event) | { | /// do something that takes some time | } | void MyClass::DoNotifyEvent2(const ONotifyEvent& event) | { | /// do something 2 | } | While within the Event1 handler, the Event2 handler will never be called. | (however an event handler in another module may be called in the mean time) | | ---- | > And it says this is archieved by interrupt masking. | I don't understand what they are trying to say. Interrupts and other lower | level operating system features are running all the time behind the scenes. | | The above rule applies to each single task (/module) using the general object | notification mechanism. There are dozens of modules/tasks running in a typical | program, and all of them run concurrently (there is a priority scheduler based | on the module's fixed priority) | | Within a single module, you can think of it as a non-multi-tasking system. If | you are writing code in multiple modules, you have to worry a little about | multi-tasking across modules, expecially if you are using shared memory | | ---- | Is there a specific scenario you are worried about ? | |
| +- 1746 RE:Member-Function atomic in OPEN-R? mcmv200i [2005/06/04-02:06] |
| | Response : Email : Edit : Delete |
|
| Mr.
AiboPet, thanks again for your answer! | Things are getting clearer and clearer... :) | | (Okay, to avoid misunderstandings: If you (and me) say "Process/Module/Task" we | mean an OPEN-R-Objekt! Right?) | | > Within your one module - there is only one execution path (one thread per | task/module) < | Oh yes, now I understand what the meant when they said "OPEN-R SDK is | single-threaded" because I wondered what kind of sense it could make to create | threads within an OPEN-R-process reacting the same way to events as the other one. | | >> And it says this is archieved by interrupt masking. | > I don't understand what they are trying to say. Interrupts and other lower | level operating system features are running all the time behind the scenes. < | | Well, me also not, that is why I was asking :) Maybe the just guessed... | | > Within a single module, you can think of it as a non-multi-tasking system. If | you are writing code in multiple modules, you have to worry a little about | multi-tasking across modules, expecially if you are using shared memory BR> | Okay, now I am getting closer to it. But there aren't any things like locks or | semaphors to protect that shared memory. Is that the reason they introduce this | Observer:Subject->Ready >> SubjectWritingToSharedMemory-> Subject | Observer->Notify >> ObserverReadingSharedMemory Stuff? | | > Is there a specific scenario you are worried about ? < | | No, there isnt't. I just want to understand the background of programming aibo | and explain it in my thesis. | (The article that 2005/06/04-02:05 posted, was modified by Author) | |
| |- 1747 RE:Member-Function atomic in OPEN-R? ejt [2005/06/04-04:45] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | >But
there aren't any things like locks or | | >semaphors to protect that shared memory. Is that the reason they introduce this | | >Observer:Subject->Ready >> SubjectWritingToSharedMemory-> Subject | | >Observer->Notify >> ObserverReadingSharedMemory Stuff? | | | | Correct, and the lack of OS-provided semaphores and locks is a real nuisance ;) | | | | Using the messages to communicate locking status is interesting -- I suppose | | that could work, but it would be a little difficult to work with since you have | | to wait for the confirmation of reception to come back before you can be sure | | you have your lock. (which implies returning out of the current callback, and | | then having to find your way back again later... I don't think you would be able | | to implement this as a pair of lock()/unlock() calls within a function.) | | | | Anyway, it is possible to write software-only mutual exclusion locks, there's a | | few research papers on it. I implemented one in Tekkotsu's MutexLock.h: | | http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/Shared/Attic/MutexLock.h?rev=HEAD&content-type=text/vnd.viewcvs-markup | | |
| | +- 1749 RE:Member-Function atomic in OPEN-R? mcmv200i [2005/06/06-19:42] |
| | | Response : Email : Edit : Delete |
|
| | Hello
Ethan! | | | | >Correct, and the lack of OS-provided semaphores and locks is a real nuisance ;) | | | | Of course , it is :) | | | | > Using the messages to communicate locking status is interesting -- I suppose | | that could work, but it would be a little difficult to work with since you have | | to wait for the confirmation of reception to come back before you can be sure | | you have your lock. < | | | | As I understand, I don't think so. Let me explain my point of view: | | | | You use a shared memory region only for one direction of communication. | | If Obersver does Observer:Subject->Ready, he know "I don't have the lock | | anymore" and then, if the subjects receives the Notify-message (The timestamp | | for that is definetaly after the timestamp for sending the msg.), the subject | | know "I have the lock now". When then the subject has finished writing the | | memory, it sends Subject Subject:Observer->Notify and knows "I don't have the | | lock anymore" before the Observer knows he has the lock. | | | | The only proplem that could occur is that a message is lost and then nobody has | | the lock (and nobody will have it again). | | | | > I don't think you would be able to implement this as a pair of lock()/unlock() | | calls within a function < | | | | I don't think so either. | | | | > Anyway, it is possible to write software-only mutual exclusion locks, there's | | a few research papers on it. I implemented one in Tekkotsu's MutexLock.h < | | | | Anyway, this is definetaly a better solution! :) | | | | Greetings | | Michael | | |
| | +- 1752 RE:Member-Function atomic in OPEN-R? ejt [2005/06/07-06:27] |
| | Response : HomePage : Email : Edit : Delete |
|
| >When
then the subject has finished writing the | >memory, it sends Subject Subject:Observer->Notify and knows "I don't have the | >lock anymore" before the Observer knows he has the lock. | | Ahh, this is mutual exclusion via token passing. The key here is that everyone | is always *giving* the lock away, as opposed to *taking* the lock only when you | need it (which was the usage I was thinking of). Simplifies communication, but | isn't as flexible in usage. | |
| +- 1748 RE:Member-Function atomic in OPEN-R? AiboPet [2005/06/06-01:38] |
| Response : Edit : Delete |
|
>(Okay,
to avoid misunderstandings: If you (and me) say "Process/Module/Task" we >mean an OPEN-R-Objekt! Right?) Correct. The Open-R "Object" is mostly a packaging for the initialization and notification (callback) mechanisms necessary for your module. Could be done in straight "C" (the glue generated by the stub tools map from standard "C" to the C++ member functions) For a more elaborate object oriented architecture, look at Tekkotsu [where C++ inheritance is used for something significant] ----- re: locks and semaphores and threads etc Some of the lower level OpenR functionality can be found in the (undocumented) system headers. You can sometimes adapt them a little to your needs. Requires some trial-and-error. However, much of the time core pieces are missing (the OpenR SDK headers are stripped down) In the easiest case, place everything in one big module if you can and manage the memory yourself. Think single threaded. |
| V 1708 Open-R SDK newbie question framebuffer [2005/05/14-13:09] |
| | Response : Edit : Delete |
|
| Hi,
I just wanted to ask can OPEN-R SDK be used with Visual Studio .NET
2003? | | Thanks in advance, | fb. | |
| +- 1711 RE:Open-R SDK newbie question AiboPet [2005/05/16-23:01] |
| | Response : HomePage : Edit : Delete |
|
| >Hi,
I just wanted to ask can OPEN-R SDK be used with Visual Studio .NET
2003? | | It is possible to hook up Visual Studio (whatever version) to edit your files | and spawn command line builds. You don't get integrated build options or any | debugging features [this is true of just about any batch build process] | | In the case of the Open-R SDK, you have to spawn 'cygwin' build - which can be | a pain to setup. | | My suggestion: use whatever editor you like, then type "make install" at the | cygwin command line to build. Visual Studio isn't adding much value in this | combination. | |
| +- 1744 RE:Open-R SDK newbie question mcmv200i [2005/06/02-17:14] |
| Response : Email : Edit : Delete |
|
>>
Hi, I just wanted to ask can OPEN-R SDK be used with Visual Studio .NET
2003? << > Visual Studio isn't adding much value in this combination. < Well, you are right! But still with Windows I use VS.NET as an editor for OPEN-R/Tekkotsu, because it has a good syntax highlighting :) (And why install another C++ editor if I have VS.NET installed?) |
| V 1741 Vote NOW for the AIBO Daft Punk Champion! openrsupport@(Administrator) [2005/05/27-02:07] |
| Response : Edit : Delete |
|
Vote NOW for the AIBO Daft Punk champion! Go to our contest website and enjoy all of the adrenaline-loaded video entries, and then cast your vote and help us find the slickest AIBO in Europe! The entries were reserved to European AIBO owners. However, anyone in the world is welcome to enjoy the videos and vote for their favorite one! Twenty-one videos are entered. It is fun to watch the videos, but please vote for your AIBO Daft Punk champion! Anyone in the world can vote! http://openr.aibo.com/openr/eng/no_perm/eventinfo.php4 Thank you, AIBO SDE Support |
| V 1737 Apertos/Aperios? OPEN-R, a complete architecture? mcmv200i [2005/05/26-23:31] |
| | Response : Email : Edit : Delete |
|
| I
realized, there are these papers "An Open Architecture for Robot | Entertainment" by Masahiro Fujita and Koji Kageyama and "The Apertos Reflective | Operating System: The Concept and Its Implementation" by Yasuhiko Yokote. They | are talking in it about a system called "Apertos". Is it the same as "Aperios" | which runs on the Aibos? And in "An Open Architecture for Robot Entertainment" | they speak about OPEN-R as a complete architecture, including the hardware. And | there are three layers (HAL, SSL, APL). Has that idea changed, is OPEN-R now | only the software part of Sony's robots? Because in the OPEN-R Programmers guide | they only talk about 2 layers ("system layer" and "application layer") which | could be possibly the upper two layers of "An Open Architecture for Robot | Entertainment" ("Application Layer" and "System Service Layer") or is the "new" | sytem layer = System Service Layer + Hardware Adaptation Layer? Are there some | new papers about OPEN-R and Aperios? For example, what scheduling strategy is used? | (The article that 2005/05/26-23:29 posted, was modified by Author) | |
| +- 1739 RE:Apertos/Aperios? OPEN-R, a complete architecture? AiboPet [2005/05/27-00:45] |
| Response : HomePage : Edit : Delete |
|
Much
of this is IIRC (If I Recall Correctly) and AFAIK (As Far As I Know). Much of the history is not described well in official documentation. >I realized, there are these papers ... They >are talking in it about a system called "Apertos". Is it the same as "Aperios" Apertos is the older name. They changed it to Aperios shortly(?) before the first Open-R press releases [Apertos/Aperios = core operating system with messaging, OpenR adds more robot-like stuff on top] I believe there was a stand-alone computer running Apertos (Japanese, found some old documents on the web a while ago) NOTE: for this discussion, the "OpenR SDK" is something else (ie. not "OpenR") > And in "An Open Architecture for Robot Entertainment" >they speak about OPEN-R as a complete architecture... Many of the concepts are still around. Reading the old papers is a good background and a little history -- HOWEVER, I wouldn't take anything verbatim. Even the two papers I've read have contradicitons [and they are both very old] The exact implementation of OpenR has changed over time (the ERS-11x to ERS- 2x0 was a big change too). > Are there some >new papers about OPEN-R and Aperios? Not AFAIK -- other than the documentation you find here [focusing on the OpenR SDK - which is only part of the story] There is a Japanese OpenR SDK book available (the English translation is still in limbo I guess) -- I don't know if it adds more. > what scheduling strategy is used? It is a somewhat unique threading/task model that is built on the object communication system [ie. not like regular threads, which is why there are no 'sleep' primitives - all your code is structured around object notification handlers as described in the SDK documentation] |
| V 1736 Robocup Walking Style on Front Knees alevy222 [2005/05/26-12:51] |
| | Response : Email : Edit : Delete |
|
| Has
anybody had any luck downloading and installing some of the Robocup
Soccer | software, such as the faster walking style, and can you incorperate it into | ERS 7 Mind 2 or do you loose all personality in your AIBO and have just a fast | walking robot. I just bought my fisrt Aibo and want to start programing it | and am looking for some resources. | | Thanks Al | |
| +- 1738 RE:Robocup Walking Style on Front Knees AiboPet [2005/05/27-00:33] |
| Response : HomePage : Edit : Delete |
|
>Has
anybody had any luck downloading and installing some of the Robocup
Soccer >software, such as the faster walking style Yes. The CMUPack (RoboCup 2002 winner) walking algorithm is the most popular. I've incorporated it into WalkBench and AiboSpeedRacer (both OpenR SDK programs, ERS-7 and ERS-2x0) - see videos too http://www.aibohack.com/openr_sdk/speedracer.htm [newer] http://www.aibohack.com/openr_sdk/walkbench.htm [older] Also the Tekkotsu framework uses it (packaged differently). It also has other parts of the RoboCup vision software. Most of the other Robocup software is not as generally useful (custom algorithms for vision on the playfield, world models of the playfield etc) ---- > and can you incorperate it into ERS 7 Mind 2 Sorry, no. > or do you loose all personality in your AIBO and have just a fast >walking robot. Yes, that describes the trade-off rather well. Also you need a "PMS" memory stick (can't use the 32MB AiboMind 1 or 2 stick) ------ The OpenR SDK creates standalone programs, very powerful but very low level. You can do almost anything you want - but you have to nearly start from scratch. You have to teach the dog how to walk. With AiboMind2 and RFW you have the other extreme. Much built-in functionality (walking, ball and bone behavior, voice recognition, visual pattern recognition, ...). Remote control type of programmability (ie. from a PC through WiFi). Can't be programmed to do brand new things. RCODE is somewhere in the middle. ===== > I just bought my fisrt Aibo and want to start programing it >and am looking for some resources. Check the resources here, the FAQs and also my advice for getting started with the OpenR SDK http://www.aibohack.com/openr_sdk/advice.htm A little out of date, but the short version is: consider using the higher level RCODE or RFW if it can solve your problems Picking which approach you want to use is not an obvious choice. The worlds of "RFW", "RCODE" and "OpenR SDK" (including Tekkotsu) have different pluses- and-minuses and relatively little sharing. Sometimes it is one essential feature that will force the decision. ie. if you *need* knee walking, then OpenR SDK (optionally with Tekkotsu) is the only option. |
| V 1705 Catching a MECHA_LOAD_HIGH Shutdown mrusso [2005/05/13-02:10] |
| | Response : Email : Edit : Delete |
|
| Hello,
I am trying to detect MECHA_LOAD_HIGH shutdown conditions before they | happen on an ERS-7. I know I have to get the pwmDuty data from the OJointValue, | but I have two problems. | | First off, how is the number stored in pwmDuty bounded? In testing, I gather | values between 1652 and -3452. But I am unsure what percentage of time ON these | numbers equate to. | | Also, what are the specs for shutdown on the ERS-7? In the FAQ the shutdown for | an ERS220 is defined as 78% max pulse width for 30 seconds. At the very least | the total time seems to be decreased as my AIBO can crash within 10 seconds of | encountering an obstacle. (I am already check for overdraws on current so I | know it is not that). | | Your help in this matter in greatly appreciated. Thank you for your time. | (The article that 2005/05/13-02:03 posted, was modified by Author) | |
| +- 1707 RE:Catching a MECHA_LOAD_HIGH Shutdown openrsupport@(Administrator) [2005/05/14-04:06] |
| | Response : Edit : Delete |
|
| >First
off, how is the number stored in pwmDuty bounded? In testing, I gather | >values between 1652 and -3452. But I am unsure what percentage of time ON these | >numbers equate to. | | The number for pwmDuty is a requested current from the PID controller. The | maximum current (pwmDuty) that can be supplied to a motor is -255 and 255. | Negative current is used to drive the motor in the opposite direction, so the | absolute value is an effective value. | | >Also, what are the specs for shutdown on the ERS-7? In the FAQ the shutdown for | >an ERS220 is defined as 78% max pulse width for 30 seconds. At the very least | >the total time seems to be decreased as my AIBO can crash within 10 seconds of | >encountering an obstacle. (I am already check for overdraws on current so I | >know it is not that). | | The MECH_LOAD_HIGH shutdown specs are the same as the ERS-200 series. | 255 * 78% = 198.9 is the threshold. | | If the time to shutdown is less than 30 seconds, the cause may be | 'BATTERY_OVER_CURRENT'. See C0010 FAQ. | https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq4.php4#C0010 | | The fact that the pwmDuty is 1652 and -3452 is not normal for AIBO. | The shutdown mechanism is the final layer of hardware protection for AIBO. | | We recommend fixing your program by preventing strenuous motions for AIBO. | | AIBO SDE Support | |
| +- 1724 RE:Catching a MECHA_LOAD_HIGH Shutdown mrusso [2005/05/19-03:49] |
| | Response : Email : Edit : Delete |
|
| >The
number for pwmDuty is a requested current from the PID controller. The | >maximum current (pwmDuty) that can be supplied to a motor is -255 and 255. | >Negative current is used to drive the motor in the opposite direction, so the | >absolute value is an effective value. | | After running my programs on six different AIBO ERS-7s and examining the pwmDuty | of all 12 leg joints on all of them, I am still getting values that are outside | of the 255 to -255 range. The AIBOs get these values during normal operation | and on programs that do not crash the robot. They crash on a MECHA_LOAD_HIGH | Shutdown when the average of the abs of the pwmDuty on a given joint is between | 400 and 600 over the past 30 seconds. | | > We recommend fixing your program by preventing strenuous motions for AIBO. | | I am trying to implement cataching a MECHA_LOAD_HIGH Shutdown so I can prevent | the AIBO from running itself to strenously. I want it to detect when its | working itself too hard and correct its behavior accordingly. However, with the | pwmDuty giving me unreliable/incorrect values, I am unsure how to proceed. | | Have other people had this problem? How has Sony dealt with this problem in the | past? Thanks for your time. | |
| |- 1734 RE:Catching a MECHA_LOAD_HIGH Shutdown mrusso [2005/05/26-03:47] |
| | Response : Edit : Delete |
|
| Anyone?
Anyone? Bueller? Anyone? | | >>The number for pwmDuty is a requested current from the PID controller. The | >>maximum current (pwmDuty) that can be supplied to a motor is -255 and 255. | >>Negative current is used to drive the motor in the opposite direction, so the | >>absolute value is an effective value. | > | >After running my programs on six different AIBO ERS-7s and examining the pwmDuty | >of all 12 leg joints on all of them, I am still getting values that are outside | >of the 255 to -255 range. The AIBOs get these values during normal operation | >and on programs that do not crash the robot. They crash on a MECHA_LOAD_HIGH | >Shutdown when the average of the abs of the pwmDuty on a given joint is between | >400 and 600 over the past 30 seconds. | > | >> We recommend fixing your program by preventing strenuous motions for AIBO. | > | >I am trying to implement cataching a MECHA_LOAD_HIGH Shutdown so I can prevent | >the AIBO from running itself to strenously. I want it to detect when its | >working itself too hard and correct its behavior accordingly. However, with the | >pwmDuty giving me unreliable/incorrect values, I am unsure how to proceed. | > | >Have other people had this problem? How has Sony dealt with this problem in the | >past? Thanks for your time. | |
| +- 1735 RE:Catching a MECHA_LOAD_HIGH Shutdown openrsupport@(Administrator) [2005/05/26-10:06] |
| Response : Edit : Delete |
|
I
apologize for the misunderstanding. pwmDuty is the requested amount of
power from the PID controller. When the PID parameters are set to be strong and the target angle is far from the current angle, the PID controller requests the maximum power. This value can be outside the bounds of +/- 255. However, the power supply is limited to +/- 255. So, if pwmDuty is greater than 255, the motors will only be supplied with 255. 255 is also an unsafe amount of power over a period of time. There is a safety mechanism that will shutdown AIBO if a motor draws 78% (198.9) power for 30 seconds. To prevent the MECHA_LOAD_HIGH emergency shutdown, you can weaken the PID parameters and also set the target joint angles closer to the current joint angles (move the joints slower). Thank you, AIBO SDE Support |
| V 1712 Connecting to multiple aibo via remote framework dbroomfield [2005/05/17-02:52] |
| | Response : Edit : Delete |
|
| has anyone done work with remote framework and multiple aibo. | | I have been testing a connection and polling msg as suggested by aibopet in my | other question and i have no problems connecting to the first aibo.. but the | second aibo wont connect. | | according to the remote framework doco you can connect to upto 16 aibo. | |
| |- 1713 RE:Connecting to multiple aibo via remote framework dbroomfield [2005/05/17-06:39] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | if it helps. | | When i connect to the second AIBO i get the error VAIBO_CONNECT_BUSY | | I have tried putting a 30 second sleep in, but that does not help. | | |
| | +- 1714 RE:Connecting to multiple aibo via remote framework TodboT [2005/05/17-08:58] |
| | | Response : Edit : Delete |
|
| | Yes,
I've connected multiple robots before. Everything worked fine. Did you | | create multiple instances of CVAIBO? | | |
| | +- 1716 RE:Connecting to multiple aibo via remote framework dbroomfield [2005/05/17-17:48] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | Yeah..
although I wasn't sure you had to do that reading though the doco. | | I basicly create an array of CVAIBO and access each one with a counter. | | | | the error I get is -23 (connect busy) | | |
| | +- 1723 RE:Connecting to multiple aibo via remote framework AiboPet [2005/05/19-00:28] |
| | | Response : HomePage : Edit : Delete |
|
| | >the
error I get is -23 (connect busy) | | | | Update (based on offline conversation): apparently the problem was related to | | not having a Windows message pump running early in the program. | | | | In order for the Connect() process to complete, Windows messages must be | | pumping (GetMessage/DispatchMessage) for the communication window. | | | | Without it, the second pAibo2->Connect() will fail because the first connect | | hasn't fully finished (even though it returned from the first pAibo1->Connect | | () ) | | |
| | +- 1725 RE:Connecting to multiple aibo via remote framework dbroomfield [2005/05/20-01:09] |
| | Response : HomePage : Email : Edit : Delete |
|
| Thanks..
I held off responding myself until I had managed to test. | | Because I am writting a console application I had to move the connect commands | into the message loop so that the message queue could be processed before | trying to connect to the second AIBO. | |
| +- 1733 RE:Connecting to multiple aibo via remote framework dbroomfield [2005/05/25-03:55] |
| Response : HomePage : Email : Edit : Delete |
|
Hello I can now connect to multiple aibo.. ironicly I found the same info repeated in a FAQ on this site today! I have a problem capturing data from multiple AIBO.. when I connect to the second AIBO and issue a APPCMD_IMAGE_PLAY the first aibo works, the second returns -31 (Command error) |
| V 1669 Which Platform to use (for research)? Tekkotsu / ARF? mcmv200i [2005/04/15-22:27] |
| | Response : Email : Edit : Delete |
|
| Hello Aibos out there, it's me again. | | As I told you, I doing a student research project with AIBO (at Dresden | Univerity of Technology). My University bought an Aibo with the main intension | to do experiments with (new) navigation and localisation algorithms (maybe | checked in parallel on other platforms as well). My assignment now is to find | out how to programm aibo, how his sensors and joints work... (a base function | check) and to programme a (simple) navigation modul which enables aibo to dodge | obstacles dynamically and to find visual landmarks. (The answer to the question | to AiboPet: I don't want to create new walk styles etc. but still have the | option to get all of the low sensor info). | | Now I want you to help me with my decision which programming platform to use. | The two frameworks which are possible to use in my opinion are the Aibo Remote | Framework and the Tekkotsu Framework. | | 1.) Do you have another idea? | 2.) Does anybody here have experimence with one or both of the frameworks? | 3.) Is it possible to control Aibo remotly with Tekkotsu Framework? (My teacher | wants to have the ability to switch between running the same programs on the PC | OR on AIBO). | 4.) Is Tekkotsu 100% compatible with OPEN-R, so can I use non Tekkotsu OPEN-R | moduls in parallel with Tekkotsu Framwork? | 5.) Which framework would you recommend? | | Many thanks in advance! | Michi | |
| |- 1670 RE:Which Platform to use (for research)? Tekkotsu / ARF? AiboPet [2005/04/16-00:16] |
| | Response : HomePage : Edit : Delete |
|
| >The
two frameworks which are possible to use in my opinion are the Aibo
Remote | >Framework and the Tekkotsu Framework. | > | >1.) Do you have another idea? | | There is also RCODE WiFi (telnet-like). Great for handling the remote control | aspects (all the sensors are available in some form, walking and skit | performance is brain-dead easy). Absolutely the easiest way of programming | AIBO, letting you focus your time on writing code on the PC. | | The only missing piece is Video capture for the ERS-7 (the ERS-2x0 models | have 'RCodePlus' which supplements RCODE with video capture). If you had an | ERS-2x0 robot, that would be my #1 suggestion. | | There is also Remote Processing OpenR - not to be confused with the Remote | Framework. This allows you to write OpenR programs that run on your PC (and | RPC to the real OpenR running on the dog) | | >3.) Is it possible to control Aibo remotly with Tekkotsu Framework? | Yes. See the website. They have a console telnet-like multipport interface, as | well as GUI controllers | | > (My teacher | >wants to have the ability to switch between running the same programs on the | PC | >OR on AIBO). | | Having the *same* program running on the AIBO and the PC -- thats a big | constraint (see below) | | >4.) Is Tekkotsu 100% compatible with OPEN-R, so can I use non Tekkotsu OPEN-R | >moduls in parallel with Tekkotsu Framwork? | | Standard (non Tekkotsu) Open-R modules must be adapted/wrapped to fit into | Tekkotsu. Fortunately for you, 'ejt' and team have done this already for the | most useful ones (ie. walking and vision from RoboCup software) | | The converse is much harder (modules written for Tekkotsu won't hook into a | standard Open-R structured app) | | >5.) Which framework would you recommend? | | Even with all the hints given, it is hard to recommend one "best" approach. | Also depends on who you ask. My advice is generally to start with the most | existing functionality like RCODE or RFW, and only go to OpenR SDK level if | absolutely needed. The Tekkotsu folks will give different advice... | | ----- | The best configuration IMHO is having the AIBO do robot stuff, and having the | PC do processing stuff (with more programming tool choices and easier | debugging). The AIBO can't do anything without the PC WiFi connection. This | works with any of the available choices. | | However if your teacher has added additional constraints: | > "(My teacher wants to have the ability to switch between running the same | programs on the PC OR on AIBO)" | | Please explain that in more detail. If that is a *requirement* (a need, not a | want), that may restrict your option to only one. | | For example: | - If you want the *same* program to run on the PC without an AIBO, then you | have to do a lot of work yourself (essentially writing a partial AIBO emulator) | - If you want the same program to run on the PC with AIBO, or with AIBO stand- | alone, there are much fewer choices (RFW and RCODE won't do it, Remote | Processing OpenR is perhaps the best choice there) | - ... | |
| |- 1671 RE:Which Platform to use (for research)? Tekkotsu / ARF? ejt [2005/04/18-04:55] |
| | Response : HomePage : Email : Edit : Delete |
|
| >1.) Do you have another idea? | Although these may violate your on-or-offboard constraint, you might want to look at these two | projects, which each have some interesting aspects: | Pyro (Python Robotics, offboard, communicating with Tekkotsu, which runs onboard) | http://emergent.brynmawr.edu/pyro/?page=Using_20the_20Sony_20AIBO_20Robot | URBI (custom language, onboard using OPEN-R SDK) | http://www.urbiforge.com/ | | >2.) Does anybody here have experimence with one or both of the frameworks? | I'll warn that I have only a overview knowledge of RFW. Since RFW code is forever stuck off-board, and | RFW is itself restricted to Windows PCs, it's not an option for our development. (We want to do as much | on-board as possible, and run on multiple platforms. Namely, I use an iBook at home, and Linux in the | lab) | Thus, Tekkotsu is basically all on-board, with PC-side monitoring tools written in Java, with ("dusty") | MATLAB hooks. | | >3.) Is it possible to control Aibo remotly with Tekkotsu Framework? (My teacher | >wants to have the ability to switch between running the same programs on the PC | >OR on AIBO). | Our PC-side API is not nearly as well organized as RFW -- if you know your code is going to run | offboard, and you're willing to give up on any onboard processing, RFW may be easier to use. | | Otherwise, PC-side control has been done with Tekkotsu, and there's several ways to approach it. | | One approach is to extend our Java tools -- we have tools for basic remote control (walk and head | pointing) as well as reception of sensor information (vision and raw sensors), which you could modify | and link together to do processing and control. You can also do combinations where some code runs | onboard, and some (say vision) runs offboard. | For instance, a group of students in a summer program here did a line follower by adding code on the | PC side to the Vision GUI tool to detect the line, and then send walking commands to the walk remote | control to follow it. This worked fairly well - the network protocol code is separated into its own class, | so you can combine multiple communication sockets in different ways, without needing the GUI. | | However, the PC-side protocol code is only in Java, so if you want to write code that at least has some | chance of running on either AIBO or PC, you'll need to port the Java networking code to C++ in order to | handle the communications. I think this may have already been done by the Pyro group. | | The third option is not yet available, but I am currently porting Tekkotsu to the PC. The ability to run in | "simulation" on logged sensor data is partially implemented, and I'm currently working on networking | support, to provide remote processing using live telemetry from the network. Interest has been | expressed by others to later add environmental simulation as well, but I cannot speak to that. | However, this is work in progress, and will not be stable for some time yet. (e.g. a month or so) | | >4.) Is Tekkotsu 100% compatible with OPEN-R, so can I use non Tekkotsu OPEN-R | >moduls in parallel with Tekkotsu Framwork? | Well, Tekkotsu is only an extension of OPEN-R, so we cannot be anything but "100% compatible", | otherwise we wouldn't run! The code is all open source, so if you want to add communication with | additional processes, you can add it as you would any other OPEN-R object. (our "main" OPEN-R object | is the MMCombo class, which is run as two copies... one declares itself "motion", and the other "main". | That's a little fancy, but each copy is just a normal OPEN-R object.) | We borrow some OPEN-R terminology and style, such as behaviors' DoStart and DoStop functions, so | it's fairly painless to port code back and forth between OPEN-R objects and Tekkotsu behaviors. (but of | course there are differences, so porting is not a null-op!) | | >5.) Which framework would you recommend? | Think carefully about your on-or-offboard requirements. | RFW is going to restrict all of your code to offboard processing, but provides a good API for doing so. | Tekkotsu is intended for onboard processing, but it is possible to do offboard as well (or in | combination), with a variety of options, but more complexity. | |
| |- 1685 RE:Which Platform to use (for research)? Tekkotsu / ARF? mcmv200i [2005/04/27-00:31] |
| | Response : Email : Edit : Delete |
|
| Thanks
a lot, AiboPet and Ethan, for your help! It was VERY usefull to me! | | I had a meeting with my teacher yesterday and we decided to use Tekkotsu. | Reasons are the Matlab and Java Interfacem and the possibility to read all the | build-in-function code when you need to do an error-search. But still ARF is for | some purpose a good option. | | Thanks again for helping me! I hope I have the time to answer you more detailled. | | Sincerely | Michael | |
| +- 1726 RE:Which Platform to use (for research)? Tekkotsu / ARF? pszu [2005/05/21-02:40] |
| | Response : Edit : Delete |
|
| Hi, | | I'm doing similar stuff - localization, landmark recognition using ARF, my | Master Thesis will (hopefully) be finished next month. I integrated ARF with | Intel's OpenCV library and wrote a module to decompress Aibo's image data on- | the-fly to IplImage* structure and I can admit this makes real-time image | processing MUCH easier. If anyone's interested and I can try to publish a small | tutorial on a webpage. | | Regards, | Przemek | | P.S. I'd like to attach front and side views of Aibo to my thesis | (ModelInformation_7_E.pdf). Is this legal (can I do it without breaking any | copyright laws?) (or whom to contact in this matter). Thans for any advice. | (The article that 2005/05/21-02:23 posted, was modified by Author) | |
| +- 1732 RE:Which Platform to use (for research)? Tekkotsu / ARF? openrsupport@(Administrator) [2005/05/25-02:56] |
| Response : Edit : Delete |
|
>P.S.
I'd like to attach front and side views of Aibo to my thesis >(ModelInformation_7_E.pdf). Is this legal (can I do it without breaking any >copyright laws?) (or whom to contact in this matter). Thans for any advice. >(The article that 2005/05/21-02:23 posted, was modified by Author) Yes! Please see FAQ ID:A0023. https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq2.php4#A0023 Thank you, AIBO SDE Support |
| V 1727 Advice needed for Website to control AIBO robot swei80 [2005/05/23-13:33] |
| | Response : Edit : Delete |
|
| HIHI, | | I am a research student and currently I am doing a research on AIBO robot. I | am required to setup a webpage for user to send command to the AIBO ERS7 | around the world and in the same time able to see the AIBO dog through the | sensors. | | Is there any advice u guys can provide me regarding the platform to use for | the webpage? | Things to take note is i will be programming some autonomous behavior into | AIBO as well and i am considering R-Code. | |
| +- 1730 RE:Advice needed for Website to control AIBO robot AiboPet [2005/05/23-23:36] |
| Response : HomePage : Edit : Delete |
|
> I >am required to setup a webpage for user to send command to the AIBO ERS7 >around the world and in the same time able to see the AIBO dog through the >sensors. This was done before on the ERS-2x0 (using RCodePlus). Using the ERS-7 RCODE will not work because there is no camera capture supported. ----- >Is there any advice u guys can provide me regarding the platform to use for >the webpage? For the ERS-7, look at "RFW". RFW does most of what you need (AIBO running AiboMind2, and you can break in and remote control him, then restore him to regular behavior). See the RFW samples - they have "NetControl" and "TakePicture" samples that do a lot of what you need -- a great place to start. They have webpages and PHP scripts. For making it public to the Internet, you will need to have a Windows PC running a webserver and connecting to the AIBO at the same time [or some other proxy mechanism] >Things to take note is i will be programming some autonomous behavior into >AIBO as well and i am considering R-Code. |
| V 1717 How are the IR sensor values discretisized? mcmv200i [2005/05/18-02:20] |
| | Response : Edit : Delete |
|
| Does
anybody (for example one of the sony members) know something about the | value discretization of the 3 IR sensors of ERS-7? What are the step values? How | many values are there (bit-resolution)? It is NOT the resolution of the data | types, it is less! | | Can anybody help me? | Thanks in advance! | |
| +- 1722 RE:How are the IR sensor values discretisized? AiboPet [2005/05/19-00:25] |
| Response : HomePage : Edit : Delete |
|
>Does
anybody (for example one of the sony members) know something about the >value discretization of the 3 IR sensors of ERS-7? What are the step values? The sensors themselves are the limiting factor Please read this excellent description of the technology used in the Sharp sensors http://www.acroname.com/robotics/info/articles/sharp/sharp.html I believe AIBO is using the "GP2D02" or very similar part [last time I checked] That part returns a single byte value in the range 10 -> 80 cm |
| V 1720 new motions in the MoNet example badia [2005/05/18-22:28] |
| | Response : Edit : Delete |
|
| I am
trying to add new motions to the MoNet example of the OPEN_R SDK. | | Can I use the ODA file generated with MEdit + RTool to do it? | | Does anybody know how to do it? | | Thanx | |
| +- 1721 RE:new motions in the MoNet example AiboPet [2005/05/19-00:21] |
| Response : HomePage : Edit : Delete |
|
>I
am trying to add new motions to the MoNet example of the OPEN_R SDK. > >Can I use the ODA file generated with MEdit + RTool to do it? Any of the MTN editors should work [I generally recommend Skitter, but MEdit or my AEditor23 will work as well] Once you get the MTN files you must combine them into an .ODA file (MOTION.ODA or MOTION7.ODA). RTool will do this. Ignore the .MWC creation feature NOTE: there are other tools like YART_APG and MAKEODA for building .ODA files This process is more-or-less the same for RCODE, RFW or Monet (OpenR-SDK sample) -- depending on your choice of editing and combining tools. ---- There is one extra step for using motions in the Monet (OpenR-SDK sample): Edit the "MONETCMD.CFG" on the memory stick to include your new motions (or motions with sound). The format is cryptic, so copy an existing motion entry and give it a new ID. NOTE: there are two separate MONETCMD.CFG files (one for ERS-7, the other for ERS-2x0) |
| V 1718 Spectral luminous efficiency function V(λ) of Aibos camera mcmv200i [2005/05/18-02:36] |
| | Response : Email : Edit : Delete |
|
| The
spectral luminous efficiency function V(λ) of the human eye
gives the | relative sensibility depending on the frequency of the light. (For example (here | also depending on the age): | http://unit.aist.go.jp/pubrel/indusstan/ljis/theme/final/finalreports/image/e-17.jpg). | Is there such a thing also availible for aibos camera and could I get it | somewhere (ERS-7)? | |
| +- 1719 RE:Spectral luminous efficiency function V(λ) of Aibos camer ejt [2005/05/18-04:55] |
| Response : HomePage : Email : Edit : Delete |
|
This
isn't exactly what you are asking, but I just happened to be working on camera calibration during my previous vision class project, which you might be interested in: http://ejt.no-ip.org/stuff/16-721/ In particular, the response function calibration which is done in the second half. Unfortunately, I only work with the intensity response of the Y channel, and essentially ignore the U and V channels, but I think you could do a similar calibration with them as well, from which you might be able to compute the wavelength sensitivities... |
| V 1691 FAQ ID:B0004 Gilles [2005/05/06-02:24] |
| | Response : Edit : Delete |
|
| I
tried to execute on my ER7-7 a motion made with Motion Editor by using
RFW, | following instructions of FAQ ID:B0004 (How do I execute the sample motions | included in the AIBO Motion Editor?) but it does'nt run : Aibo stops | immediatly when I push the "contents renewal button". | Someone has an idea ? | |
| +- 1692 RE:FAQ ID:B0004 TodboT [2005/05/06-09:32] |
| | Response : Edit : Delete |
|
| I
forgot that you could do this with remote framework. | | I followed the steps (except for 14, now Mind2 MS is sued) and everything worked | fine. Does your AEP work? If so, then I'd suggest to try again, and make sure | you follow the steps correctly. | | TodboT | |
| |- 1694 RE:FAQ ID:B0004 Gilou94 [2005/05/06-14:05] |
| | | Response : Edit : Delete |
|
| | In
fact, I have an error message in step 13 ("Make Files" in Motion
Editor) : something like "error sum". | | But the files (.cfg and .oda) are created yet. I don't know why for instance. | | My AEP works fine. | | Subsidiary question : is it possible to add motions made with ME into AEP motions folder ? | | |
| | +- 1715 RE:FAQ ID:B0004 TodboT [2005/05/17-09:47] |
| | Response : Edit : Delete |
|
| Make
sure that your paths are correct in the Contents Control Property window | and that MWC.CFG has the correct motion/LED/sound file names (w/o extensions). | How large are the ODA files and ERS-7.MWC? I followed the steps in the FAQ and | had the following files/sizes: | | ERS-7.MWC 154 Bytes | NETAUDIO.ODA 3.07 KB | NEDLED.ODA 12.6 KB | NEDMTN.ODA 1.62 KB | |
| +- 1696 RE:FAQ ID:B0004 openrsupport@(Administrator) [2005/05/10-10:02] |
| Response : Edit : Delete |
|
Thank
you for noticing the error in the FAQ. Steps 14 and 15 have been corrected. AIBO SDE Support |
| V 1709 Download cannot open module file: /MS/..../OBJECT_NAME.BIN aibourjc [2005/05/16-20:00] |
| | Response : Edit : Delete |
|
| Hi, | i'm trying to debug an OPEN-R code but i get that message: | | Download cannot open module file: /MS/..../OBJECT_NAME.BIN | R7KDebugAddressManager::DeleteRegion: meta object's request is not in normal | memory, address[0] | R7KDebugAddressManager::DeleteRegion: meta object's request is not in normal | memory, address[0] | [oid:80000031,prio:1] OObjectRegistry::LoadObjects() : | Load(/MS/OPEN-R/MW/OBJS/OBJECT_NAME.BIN) error 90 | | I've that .bin in the memory stick so i don't know what the problem is. | (The article that 2005/05/16-18:30 posted, was modified by Author) | |
| +- 1710 RE:Download cannot open module file: /MS/..../OBJECT_NAME.BIN AiboPet [2005/05/16-22:58] |
| Response : HomePage : Edit : Delete |
|
>Download
cannot open module file: /MS/..../OBJECT_NAME.BIN I assume that is your own module. Has it worked before ? if so have you changed anything ? ---- I suggest keeping module filenames short - no more than 8 characters This has caused problems in the past - I don't know if the ERS-7 has the same problems - but it is easy to rename it to something shorter (OBJ_NAME.BIN) just to be safe. |
| V 1706 A newbie quesion framebuffer [2005/05/13-21:03] |
| Response : Edit : Delete |
|
Hi, I am recently beginning to like programming robots and wanted to start by owning an aibo. I have previous programming experience in real-time computer graphics. I was just wondering whether there is a scope of earning from this knowledge apart for it being fun. Do other companies hire people to program on the sony platform? Thanks in advance, fb. |
| V 1698 RFW - Timeout issue dbroomfield [2005/05/10-17:07] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hello All, | | I have written a simple remote framework application.. it doesnt actually do | much yet, just connects to AIBO instructs AIBO to broadcast images and just | writes a "." to the console when the WM_IMAGETP_UPDATE event is hit. | | This is an experiment I will take further once I am happy things are working | as they should. The thing I noticed was that after two minutes AIBO | disconnects (I have a trap on the WM_VAIBO_DISCONNECT event too. | | Is there a reason for this? I rekon it's some sort of timeout, if so how do I | stop the timeout occurring? | |
| +- 1704 RE:RFW - Timeout issue AiboPet [2005/05/12-01:16] |
| Response : HomePage : Edit : Delete |
|
>I
have written a simple remote framework application.. it doesnt actually
do >much yet, just connects to AIBO instructs AIBO to broadcast images and just >writes a "." to the console when the WM_IMAGETP_UPDATE event is hit. Try adding something like "myAiboPointer->ImageGetData(...)" inside the handler for the message/event, and ignore the data. Advice/Guesses: Assuming it is a reasonably consistent 2 minute timeout, I am guessing AIBO is timing out because it isn't seeing any messages for you for 2 minutes [after initially setup]. The Image data is coming over a UDP connection. If the timeout time is less consistent (ie. not reproducable within 20 seconds), I would suspect it is a Wifi connection drop. |
| V 1690 Identify color through Camera Sensor then output it with LED sandytock [2005/05/05-11:00] |
| | Response : Email : Edit : Delete |
|
| Hi everyone, | I am trying to implement some modules that will allow the AIBO to recognize | color with camera sensor and then output the wordings of the color with its | LED. | | I have been experimenting with some of the tutorials but still progress is | slow. | Anyone did something similar before? Maybe we can discuss about it. Or is | this already implemented? | | Please do HELP~~ | | Regards~ | Sandy | |
| +- 1695 RE:Identify color through Camera Sensor then output it with LED AiboPet [2005/05/07-00:34] |
| | Response : HomePage : Edit : Delete |
|
| >I
am trying to implement some modules that will allow the AIBO to
recognize | >color with camera sensor and then output the wordings of the color with its | >LED. | | ? class project ? | | Please see one reply here: | http://bbs.aiboworld.co.uk/index.cgi?read=23990 | | Short answer: the ERS-7 LEDs can't display words - only a few limited shapes | and/or color patterns | |
| +- 1697 RE:Identify color through Camera Sensor then output it with LED swei80 [2005/05/10-10:39] |
| | Response : Email : Edit : Delete |
|
| Hihi, | I am doing this as my personal interest as well. If you guys can offer any | help~ I be most thankful for it~ | | Btw i have used skitter program to modidy the LED lights. However, juz like to | enquire how to merge the skitter program to the OPEN-R "ImageObersver and | BallTracking" module??? | | >>I am trying to implement some modules that will allow the AIBO to recognize | >>color with camera sensor and then output the wordings of the color with its | >>LED. | > | >? class project ? | > | >Please see one reply here: | >http://bbs.aiboworld.co.uk/index.cgi?read=23990 | > | >Short answer: the ERS-7 LEDs can't display words - only a few limited shapes | >and/or color patterns | |
| +- 1703 RE:Identify color through Camera Sensor then output it with LED AiboPet [2005/05/12-01:10] |
| Response : HomePage : Edit : Delete |
|
>Btw
i have used skitter program to modidy the LED lights. However, juz like
to >enquire how to merge the skitter program to the OPEN-R "ImageObersver and >BallTracking" module??? Skitter creates a .LED file which can be very easily played back using RCODE. It is not directly supported with the OpenR SDK. The .LED file is a Midi format file that contains timing information for when each LED should be turned on or off (it is possible to write a player module in OpenR, but IMHO not worth the effort unless you want long flashing LED sequences) ---- My advice: For infrequent turning on/off LEDs using the OpenR SDK, I suggest using Skitter as an exploration tool only. First learn the limitations of the LED face array using Skitter. Once you have found the combination of LEDs you like, you can control them directly in a manner similar to the "BlinkingLED7" sample (merge the code into your own OpenR program) The names of the LED primitives (eg: "PRM:/r1/c1/c2/c3/la-LED3:la") and where they are located in the faceplate are in the documentation. Skitter uses less cryptic names |
| V 1699 Yart issue swei80 [2005/05/11-01:50] |
| | Response : Email : Edit : Delete |
|
| Hi everyone, | | I am using Yart to generate certain commands for my ERS-7 AIBO. | However after i save the program in my memory stick, when i finish booting up | my AIBO, its seems to jam. Is it normal? | | Btw i copy the yart program into my memory stick then after manipulating the | program i save it by pressing the "Save RCode to Memory Stick button" | | Is this the right procedure?? | Thanks | |
| +- 1702 RE:Yart issue AiboPet [2005/05/12-01:02] |
| Response : HomePage : Edit : Delete |
|
>I
am using Yart to generate certain commands for my ERS-7 AIBO. See other thread. A "PMS" stick is needed for RCODE/YART http://www.aibohack.com/ers7/yart7.htm http://www.aibohack.com/pms.htm |
| V 1700 AIBO MIND MEMORY STICK vs AIBO PROGRAMMING STICK swei80 [2005/05/11-15:47] |
| | Response : Email : Edit : Delete |
|
| Hi,
just like to check can we do coding in AIBO MIND MEMORY STICK using
OPEN-R | or R-CODE? Cos i am using AIBO MIND MEMORY STICK rather than AIBO programming | stick and nothing seems to work for me including the sample codes. | | Btw wats the difference of it? | | Is there anyway to go around it? Please response asap as this is urgent~ | Thanks | Yee Sern Wei | |
| +- 1701 RE:AIBO MIND MEMORY STICK vs AIBO PROGRAMMING STICK AiboPet [2005/05/12-01:00] |
| Response : HomePage : Edit : Delete |
|
>Hi,
just like to check can we do coding in AIBO MIND MEMORY STICK using
OPEN-R >or R-CODE? No, you need a PMS stick (AKA "RCODE stick") See the FAQ or http://www.aibohack.com/pms.htm The 32MB AiboMind stick can only run the AiboMind 1 or AiboMind 2 software. AiboMind 2 software is necessary for "RFW" |
| V 1689 How to get Aibo Status randeep [2005/04/29-14:12] |
| | Response : Edit : Delete |
|
| Hi, | | How can I know the following status of Aibo in RFW: | | 1. Aibo is doing a dance (PlayMotion): I want to let it finish before giving | the Stop command. | 2. Is there any timings defined for PlayMotion routines, for e.g.: This dance | will take 5 sec to finish. | 3. Aibo Posture: So that I can send suitable PlayMotion command. | 4. Is there a way to interpret Aibo joint values. How? | 5. Dynamic Posture: Walking etc. | | Thanks | |
| +- 1693 RE:How to get Aibo Status TodboT [2005/05/06-09:54] |
| Response : Edit : Delete |
|
Take
a look at the documentation and the sample programs. That should help
you out a lot. >1. Aibo is doing a dance (PlayMotion): I want to let it finish before giving >the Stop command. I think you only need to send a stop command if it's an infinity (repeating) motion. There's a wait flag when you call PlayMotion (see VAIBOClient.txt). If it's set, then the function doesn't return until the motion is complete. If you don't use the wait flag, then you'll need to check the semantics data for when the motion completes (BaseClient sample). >2. Is there any timings defined for PlayMotion routines, for e.g.: This dance >will take 5 sec to finish. I'm not sure. But I don't think so. >3. Aibo Posture: So that I can send suitable PlayMotion command. Use GetPosture. See VAIBOClient.txt documentation. >4. Is there a way to interpret Aibo joint values. How? Do you want to read the joint angles? If so, take a look at the GetSensorInfo sample. >5. Dynamic Posture: Walking etc. I'm not sure what you're asking, but if you want to know the posture, use GetPosture and look at VAIBODef.h to see the list of postures. When you call DoWalking, you can specify different walking patterns, but they're all built-in. No custom walking gaits. TodboT |
| V 1683 Who manage events pakitoMR [2005/04/26-21:07] |
| | Response : HomePage : Email : Edit : Delete |
|
| When
an Open-R communication occurs, the subject stores the data to send in a | shared memory area. Then, the subject notifies (Notify event) to observers. This | event is received by the observers. The question is: Who carry this event to | observers? The subject? An Open-R thread? | | Thanks in advance, | Paco. | (The article that 2005/04/26-21:03 posted, was modified by Author) | |
| +- 1688 RE:Who manage events AiboPet [2005/04/29-02:14] |
| Response : HomePage : Edit : Delete |
|
>When
an Open-R communication occurs, the subject stores the data to send in a >shared memory area. Then, the subject notifies (Notify event) to observers. This >event is received by the observers. The question is: Who carry this event to >observers? The subject? An Open-R thread? If I understand your question, correctly, it is best described as the "observer thread" (different from the "subject thread"). All the threads/tasks are managed by OpenR of course. The normal notify operation is like posting a message to another process - it will received it some time later in the normal processing order. |
| V 1684 Does SetData copy data? pakitoMR [2005/04/26-23:31] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
have a Class P. Then, I create a P instance (c++ object) called p. This | instance p has data that dinamically changes its size (a dynamic list or | something like this). It not possible to determine a priori the p size at runtime. | | I can understand how this code can work: | | subject[sbjSetUpdate]->SetData(&p, sizeof(p)); | | Is sizeof() able to determine an object size with such characteristics (dinamic | memory data size)? How it does? SetData performs a copy (memcpy) or it uses | another mechanism? | | Thanks in advance, | Paco. | |
| +- 1687 RE:Does SetData copy data? AiboPet [2005/04/29-02:09] |
| Response : HomePage : Edit : Delete |
|
>I
have a Class P. Then, I create a P instance (c++ object) called p. This >instance p has data that dinamically changes its size (a dynamic list or >something like this). It not possible to determine a priori the p size at runtime. I assume this dynamic data is member variables inside the class, eg: class P { int* dataPtr; // data is allocated with 'new' ... }; >subject[sbjSetUpdate]->SetData(&p, sizeof(p)); >Is sizeof() able to determine an object size with such characteristics This will not work. It will send the entire "p" object, but any data pointers inside the object will not make sense to another process receiving the data [data is normally not allocated in shared memory] I assume you are writing both the sending and receiving objects. I suggest either merging them into the same process (if that makes sense), or when sending the data you will have to serialize all the essential data into on linear data block (then pass it to SetData) ---- NOTE: this is a general problem in any system that much send data between processes or through RPC mechanisms - not unique to AIBO. BTW: Most versions of SetData you will use do copy the data for you (which is why it must be in a contiguous chunk of memory). There are more advanced variations that use shared memory. |
| V 1686 AIBO Remote Framework tutorials are available openrsupport@(Administrator) [2005/04/27-11:07] |
| Response : Edit : Delete |
|
You can enjoy AIBO Remote Framework by using the sample programs. A development environment (VC++) is not needed. No programming is required to run the sample programs. https://openr.aibo.com/openr/eng/perm/university2.php4 1. Trying AIBO Remote Framework 2. Make AIBO Sing a Song 3. Control AIBO and Make AIBO Self-charge 4. Using Recognition Functions with RemoteTest Thank you, AIBO SDE Support (The article that 2005/04/27-11:05 posted, was modified by Author) |
| V 1659 Aibo Remote Framework - Reading sensor information mcmv200i [2005/04/07-00:03] |
| | Response : Email : Edit : Delete |
|
| Hello aibophils :) | | I would like to ask you for help with aibo remote framework. | | I am working on a student research project with AIBO ERS-7 and for the | beginning I would like to programm a remote sensor control tool (using aibo | remote framework). | | I started using the GetSensorInfo-Example but still I have some problems on | how the sensor information is passed. Okay, this GetSensorInfo-Example reads | the Sensor Info given with the WM_VAIBO_SENSOR_DATA window message. The sensor | data is put in lParam as an array of SensorRecs. Which is | this kind of type: | | (from VAIBODEF.H) | | typedef struct SensorRec { | int sensorID; | unsigned long value; | long percentage_x; | long percentage_y; | } SensorRec, *SensorRecP; | | with (from CPCINFO.H) | | enum SensorID { | Acc1 = 0, | Acc2, Acc3, HeadTilt, HeadPan, HeadRoll, | Head1, Head2, Chin, PSD, LFJ1, LFJ2, LFJ3, LFSW, | LRJ1, LRJ2, LRJ3, LRSW, RFJ1, RFJ2, RFJ3, RFSW, | RRJ1, RRJ2, RRJ3, RRSW, BackSW, | LEar, REar, Mouth, Tail1, Tail2, Temp, | MultiSW1, MultiSW2, MultiSW3, PSDWithXY, | HeadTilt2, BackSW2, BackSW3, | PSDWithXY_FN, PSDWithXY_N, PSDWithXY_F, | PrimitiveMax | }; | | I still don't know what some of these identifiers mean. I know a lot but not | everything: | | (Okay, they said HeadRoll doesn't mean anything for ERS-7 and doesn't work. | Thats okay. But:) | | *What does MultiSW1, MultiSW2, MultiSW3 mean? | | *I guess PSDWithXY_N, PSDWithXY_F mean Face distance sensor near and far, okay. | PSDWithXY_FN must be a kind of combination of it (the real distance?). But | what is PSDWithXY then? PSD doesn't work at all. | | * Is Temp the temperature sensor? | | * What is PrimitiveMax? | | *And how can I read the information of the chest distance sensor? | | Unfortunately the documentation isnt't that comfortable to read to get this | out. Could you please give me any advice? | | Another question: Do you recommend aibo remote framework for research purpose? | Or should I better write all my programs in open-r? My idea was to write a | remote sensor control program which shows all the sensor data to the user. But | if I write my own programm with OPEN-R I can't use it because aibo remote | framework works only with aibo mind 2, right? | | I would hope for an answer of you :) | | Thanks! | | Michi | | PS: Please forgive my bad english, I am not a native speaker. | (The article that 2005/04/06-23:04 posted, was modified by Author) | |
| |- 1660 RE:Aibo Remote Framework - Reading sensor information mcmv200i [2005/04/08-23:12] |
| | | Response : Email : Edit : Delete |
|
| | There
is no answer from you, that's the reason I have to anser myself (what I | | found out) :) | | | | >*What does MultiSW1, MultiSW2, MultiSW3 mean? | | | | Don't know but there is no information passed at all (in this version) | | | | >*I guess PSDWithXY_N, PSDWithXY_F mean Face distance sensor near and far, | | okay. | | >PSDWithXY_FN must be a kind of combination of it (the real distance?). But | | >what is PSDWithXY then? PSD doesn't work at all. | | | | PSD doesn't work at all. PSDWidthXY == PSDWidthXY_N. | | N means Near sensor and F far sensor. | | | | PSDWidthXY_FN IS a combination. It has got a build-in hysteresis: | | | | If PSDWidthXY_N and PSDWidthXY_F are high: PSDWidthXY_FN := PSDWidthXY_F | | If PSDWidthXY_N and PSDWidthXY_F are low: PSDWidthXY_FN := PSDWidthXY_N | | | | If the values are high and falling, PSDWidthXY_FN switches to PSDWidthXY_N, if | | PSDWidthXY_N falls below 20 cm. | | If the values are low and rising, PSDWidthXY_FN switches to PSDWidthXY_F, if | | PSDWidth_F exceeds 22 cm. | | | | >* Is Temp the temperature sensor? | | | | MIGHT be but doesn't work right now. | | | | >* What is PrimitiveMax? | | | | No idea, doesn't work. | | | | >*And how can I read the information of the chest distance sensor? | | | | You can't with ARF, as I think. | | | | >Another question: Do you recommend aibo remote framework for research purpose? | | | | Well, not really. For some special problems, maybe. But for deeper research, | | use OPEN-R (and maybe some enhancement tools/frameworks) | | |
| | +- 1661 RE:Aibo Remote Framework - Reading sensor information AiboPet [2005/04/09-01:00] |
| | | Response : HomePage : Edit : Delete |
|
| | >There
is no answer from you, that's the reason I have to anser myself (what I | | >found out) :) | | | | That's my general advice - run the program, press the sensors on the dog and | | see what bit values are returned. These are low level sensor information. | | | | As you have noticed documentation is sparse, especially on these very low | | level details. | | | | For many of the important things, there are higher level "SEMID" semantic | | messages/notifications. Wherever possible, I would recommend using them (see | | Semantics.txt). | | They are the results of the dog's logic to turn low level events into | | something more significant. For example the low level PSD info into a high | | level "SEMID_CLIFF" message when the dog is approaching a cliff (instead of | | SEMID_OBJECT or SEMID_OBSTACLE or SEMID_HAND) | | They are somewhat abstracted, so you may lose some detail on the way. | | | | >>* Is Temp the temperature sensor? | | > | | >MIGHT be but doesn't work right now. | | Try "SEMID_INTEMP". IIRC it should work. | | | | >>* What is PrimitiveMax? | | >No idea, doesn't work. | | Common "C"/"C++" trick - the last value in the enum which is useful for | | dimensioning arrays -- don't use it directly. | | | | >>*And how can I read the information of the chest distance sensor? | | See above. "GetInputSemantics" should give you most of the things you need. | | | | >>Another question: Do you recommend aibo remote framework for research | | purpose? | | >Well, not really. For some special problems, maybe. But for deeper research, | | >use OPEN-R (and maybe some enhancement tools/frameworks) | | | | It depends what you are researching. | | If you are researching new ways of walking, or new ways of how the dog reacts | | to buttons presses, then using the lower level alternatives like the OpenR SDK | | or the low level sensor info is better. | | | | If however you are happy with the normal AIBO behavior on areas you aren't | | researching (ie. you want him to walk and react like a normal AIBO most of the | | time), then RFW is the thing to do. Then you can focus on your research | | instead of having to re-invent walking or how button presses are detected. | | | | ---- | | Bottom line: try the SemanticInput approach and see if it gives you more | | satisfying experience | | [you may come to the same conclusion after all, depending on the kind of | | things you are researching] | | |
| | +- 1663 RE:Aibo Remote Framework - Reading sensor information mcmv200i [2005/04/11-18:52] |
| | | Response : Email : Edit : Delete |
|
| | Thanks,
AiboPet, for your nice and detailed answer! | | | | >For many of the important things, there are higher level "SEMID" semantic | | >messages/notifications. Wherever possible, I would recommend using them (see | | >Semantics.txt). | | >They are the results of the dog's logic to turn low level events into | | >something more significant. For example the low level PSD info into a high | | >level "SEMID_CLIFF" message when the dog is approaching a cliff (instead of | | >SEMID_OBJECT or SEMID_OBSTACLE or SEMID_HAND) | | >They are somewhat abstracted, so you may lose some detail on the way. | | | | Hey, thats a good tip. I didn't know these semantics also include obstacle | | detection data. For some research intention these might be a very good thing to | | use. But there is a problem: There is not so information on how these | | implemented things on Aibo work. So fault search could be difficult if you use | | these build-in functions of aibo. But you're right: If you are researching on a | | very high level and you do not have to know at all how obstacles are detected | | for instance ARF IS a very good thing! | | | | >>>* What is PrimitiveMax? | | >>No idea, doesn't work. | | >Common "C"/"C++" trick - the last value in the enum which is useful for | | >dimensioning arrays -- don't use it directly. | | | | Oh, good to know :) | | Thanks! | | | | >If however you are happy with the normal AIBO behavior on areas you aren't | | >researching (ie. you want him to walk and react like a normal AIBO most of the | | >time), then RFW is the thing to do. Then you can focus on your research | | >instead of having to re-invent walking or how button presses are detected. | | | | Yes, you are right! I don't want to - and with my knowledge I am not able to - | | teach Aibo walking again (for example). So ARF is an alternative. I will check | | also what enhancement frameworks build up on OPEN-R can bring. | | | | It would be good if there was the possibility to choose the level of abstraction | | (if you want raw sensor data or build-in functions e.g.) in every line of the | | code you write. Well, there might be some reasons why Sony doesn't make this | | possible :) | | | | See you | | Mike | | |
| | +- 1667 RE:Aibo Remote Framework - Reading sensor information AiboPet [2005/04/14-01:18] |
| | Response : HomePage : Edit : Delete |
|
| [selective
replies , multiple sub-thread] | | >It would be good if there was the possibility to choose the level of | abstraction | >(if you want raw sensor data or build-in functions e.g.) in every line of the | >code you write. Well, there might be some reasons why Sony doesn't make this | >possible :) | | The lower level sensor information as well as the Semantic Events are both | sent (not always together, sometimes only one). My advice is to find the right | mix that gives you the detailed info you need. Requires some trial-and-error. | | --- | > But: I get the information only when Aibo is in autonomous mode. If I set him | > remotely controlled he doesn't get any sematics data. | I had that problem at one point, not sure what I did to fix it (lately I've | been trying to go around the VAIBOSERVER.DLL hooks) | | ---- | > Can someone please tell me when the SEMID_(whatever) is sent... | | I'd like to see more detailed documentation too, or ideally open source code | to the RFW glue DLL/VAiboServer process and perhaps even an OpenR SDK - | compatible work-alike. | | I suggest asking the Sony folks for more documentation explicitly in another | thread. The audience for RFW has been small (unfortunately). Without more | people asking for things, I'm worried they may never update it. | |
| |- 1664 Receiving Semantics - Only in autonomous mode? mcmv200i [2005/04/12-17:27] |
| | | Response : Email : Edit : Delete |
|
| | Hello, | | | | thanks to AiboPet I started experimenting with semantic information. | | (Unfortunately documentation is rare here as well). But with some examples I | | found out how to use it. | | | | But: I get the information only when Aibo is in autonomous mode. If I set him | | remotely controlled he doesn't get any sematics data. So I can't use them for | | my programs? Or is it possible to use them, but I have to change something? | | (The article that 2005/04/12-01:10 posted, was modified by Author) | | |
| | +- 1682 RE:Receiving Semantics - Only in autonomous mode? openrsupport@(Administrator) [2005/04/26-03:43] |
| | Response : Edit : Delete |
|
| You
needed to enable each recognition engine after switching to remote mode. | Please see RemoteTest_E.txt in the sample source archive. | [To enable AIBO's recognition engines] | | AIBO SDE Support | |
| |- 1666 SEMID_PSDSPOT? mcmv200i [2005/04/14-00:05] |
| | Response : Email : Edit : Delete |
|
| Can
someone please tell me when the SEMID_PSDSPOT is released? Only if an
object | is recognized? | |
| +- 1668 Fall down protection works pretty well mcmv200i [2005/04/14-02:13] |
| Response : Email : Edit : Delete |
|
I
made some experiments letting aibo walk towards a cliff and I just
wanted you to know the cliff protection of aibo works pretty well!!! WOW!! (The article that 2005/04/14-02:12 posted, was modified by Author) |
| V 1678 Link problem on R-CODE SDK Downloads Page atate [2005/04/22-23:29] |
| | Response : Email : Edit : Delete |
|
| Sony,
please note a URL on the SDK downloads page for the ERS7 3D Models is | wrong. The target is ERS7-3Dmodel-r1.tar.gz not ERS7-3Dmodels-r1.tar.gz | | The ERS200 file is correctly named models. | | The current link fails after accepting the licence. | (The article that 2005/04/22-23:29 posted, was modified by Author) | |
| +- 1681 RE:Link problem on R-CODE SDK Downloads Page openrsupport@(Administrator) [2005/04/25-09:39] |
| Response : Edit : Delete |
|
Thank
you for noticing the missing link. The link has been corrected. Thank you. AIBO SDE Support >Sony, please note a URL on the SDK downloads page for the ERS7 3D Models is >wrong. The target is ERS7-3Dmodel-r1.tar.gz not ERS7-3Dmodels-r1.tar.gz > >The ERS200 file is correctly named models. > >The current link fails after accepting the licence. >(The article that 2005/04/22-23:29 posted, was modified by Author) |
| V 1672 Pink Object: Search and Track randeep [2005/04/18-19:30] |
| | Response : Edit : Delete |
|
| Hi, | | I am using RFW and asking Aibo to search and track Pink object. How can I know | that the Aibo has found a pink (or lost a previously found pink) object | programatically? | | Thanks | |
| +- 1673 RE:Pink Object: Search and Track AiboPet [2005/04/19-00:38] |
| | Response : HomePage : Edit : Delete |
|
| >Hi, | > | >I am using RFW and asking Aibo to search and track Pink object. How can I know | >that the Aibo has found a pink (or lost a previously found pink) object | >programatically? | | Semantic Event "SEMID_PINK" | | Look at the extra info (3rd arg to GetInputSemantics) | The first extra arg is either 1 (found) or 2 (lost) | | NOTE: not specific to searching/tracking. If AIBO comes across something pink, | or you shove the pink ball in his face, the SEMID_PINK will be triggered. | |
| +- 1675 RE:Pink Object: Search and Track randeep [2005/04/21-18:49] |
| | Response : Edit : Delete |
|
| Thanks. | Is it possible to get the "bounding box" of the pink pixels as well? | | >>Hi, | >> | >>I am using RFW and asking Aibo to search and track Pink object. How can I | know | >>that the Aibo has found a pink (or lost a previously found pink) object | >>programatically? | > | >Semantic Event "SEMID_PINK" | > | >Look at the extra info (3rd arg to GetInputSemantics) | >The first extra arg is either 1 (found) or 2 (lost) | > | >NOTE: not specific to searching/tracking. If AIBO comes across something pink, | >or you shove the pink ball in his face, the SEMID_PINK will be triggered. | |
| +- 1680 RE:Pink Object: Search and Track AiboPet [2005/04/24-01:18] |
| Response : HomePage : Edit : Delete |
|
>Is
it possible to get the "bounding box" of the pink pixels as well? Not with SEMID_PINK AFAIK Some of the other SEMID_s have a calculated distance to object (see Semantics.txt documentation). That calculated distance is using the number of pink pixels &/or the size of the bounding box to guess at the distance. If you are trying to process lower level camera, then you should be grabbing the camera video frame (supported in RFW as a JPG image) and do a little pixel processing yourself. Remember: RFW is a layer at the level the normal AiboMind2 software interprets things, exposed to support the AEP [ie. if you can't find a similar function in AEP, RFW may not provide that info] |
| V 1623 AIBO ERS7M2 simulator Iaia [2005/02/14-16:52] |
| | Response : Email : Edit : Delete |
|
| Hi! I'm Nadia, | I study Computer Science and I want prepare my thesis about AIBO but my | University has not still received it. There are tools to simulate the AIBO | actions? | Thank you in advance | Nadia | |
| |- 1636 RE:AIBO ERS7M2 simulator TodboT [2005/03/03-10:42] |
| | Response : Edit : Delete |
|
| I'm
not aware of a real simulator, but if you just want to create AIBO
motions | (MTN files) you can use MEdit (which you can download from this site) or Skitter | (http://www.dogsbodynet.com/skitter.html). I'd recommend Skitter. | |
| +- 1642 RE:AIBO ERS7M2 simulator outsider [2005/03/12-04:26] |
| | Response : HomePage : Edit : Delete |
|
| >Hi! I'm Nadia, | > I study Computer Science and I want prepare my thesis about AIBO but my | >University has not still received it. There are tools to simulate the AIBO | >actions? | >Thank you in advance | > Nadia | | Yes you can simulate Aibo by using the Webots comercial software | (www.cyberbotics.com). Aibo ERS-210 is fully available, and ERS-7 is almost | completed and you can simulate its dynamics (I'm working on it!!!). You can even | transfer your Webots control programs to OPEN-R directly executable under the | real robot. | |
| +- 1679 RE:AIBO ERS7M2 simulator atate [2005/04/22-23:33] |
| Response : Email : Edit : Delete |
|
I
was just looking at the Sony AIBO support in Webots MSc project page at http://poskria4.blogspot.com/ Is there any Java client program support in your simulator or plans for it? |
| V 1607 list of r-code commands?? teamdax [2005/01/29-12:50] |
| | Response : Email : Edit : Delete |
|
| is there a list of all the r-code commands? | | thanks | | TD | |
| |- 1608 RE:list of r-code commands?? AiboPet [2005/01/30-04:38] |
| | Response : HomePage : Edit : Delete |
|
| >is
there a list of all the r-code commands? | | Three different lists: | | Download the "R-CODE 110/111 and R-CODE 1.2(for ERS-210/220/310) main archive" | and it has the reference for RCODE 11x and RCODE 1.2 (rcode12ref.txt) | | Download the "R-CODE SDK main archive for ERS-7" and it has the reference for | RCODE ERS-7 (rcode-ers7-cmdref...*.txt) | | Each version has several additional files describing the built-in actions, | tone and tune detection as well as other things. | |
| +- 1634 RE:list of r-code commands?? outsider [2005/03/01-21:37] |
| | Response : Edit : Delete |
|
| >is
there a list of all the r-code commands? | > | >thanks | > | >TD | | I created a list from the R-Code manuals. You can find it here: | http://www.ouroboros.org/rcode_tutorial_1v2.pdf | http://www.ouroboros.org/aibo.html | |
| +- 1677 RE:list of r-code commands?? teamdax [2005/04/22-06:45] |
| Response : Edit : Delete |
|
>I
created a list from the R-Code manuals. You can find it here: >http://www.ouroboros.org/rcode_tutorial_1v2.pdf >http://www.ouroboros.org/aibo.html cool thanks- very useful |
| V 1676 Take the AIBO Daft Punk Challenge openrsupport@(Administrator) [2005/04/21-21:15] |
| Response : Edit : Delete |
|
Take the AIBO Daft Punk Challenge It's no secret that AIBO likes to dance but just how good is your fun-loving companion? Find out by entering the world's first ever AIBO Dance Contest! Simply program your AIBO with a freaky dance routine to Daft Punk's new single "Technologic" and send us a video. This contest is open to residents in a selection of European countries. https://openr.aibo.com/openr/eng/perm/medit/eventinfo.php4 AIBO SDE support |
| V 1674 Streaming data problem.. laue [2005/04/20-13:58] |
| Response : Edit : Delete |
|
I
finally was able to create my first program in OPEN-R!! YAY!!!! It's a simple little app that takes messages from a server and filters out the irrelevant data and only present what is needed. (Simple client/server app) My problem is this: the server has a large amount of data that it is sending back. When my Aibo receives the messages, it tends to drop parts of the message. (It's asynchronous as the server doesn't wait for a response once the send command is made by the client, so the data keeps getting sent). In my client app, I have a ReceiveCont function that calls the Receive function. It does this until the TCP status is NOTCONNECTED. While this does get most of the data, it doesn't get all of it. How can I change teh client program so that it picks up all the messages the server sends? Thanks |
| V 1638 Dynamical data of the ERS-7 Sebas [2005/03/04-19:43] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hello all, | | I am a student and I am working on a dynamical model of the ERS-7 for the Dutch | Aibo Team in The Netherlands. But to make the model realistic I need data of | each link in the ERS-7, like masses, center of mass and inertias. | | Are there people who have the same request? | Does somebody know where to ask this with Sony? | Or is there somebody who can help me with obtaining this data? | | Thanks. | | Best regards, Sebas | |
| +- 1641 RE:Dynamical data of the ERS-7 ejt [2005/03/07-02:52] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
second this request as well. This information is a necessity for cool
demos like balancing and fine | manipulation that would really make the AIBO look good and show off what it's capable of. | |
| +- 1665 RE:Dynamical data of the ERS-7 openrsupport@(Administrator) [2005/04/13-01:28] |
| Response : Edit : Delete |
|
Thank
you for showing an interest in developing new software for AIBO. Unfortunately we're currently unable to provide the data. Our FAQ has been updated to answer your request. https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq2.php4#A0048 ID:A0048 Q. I need certain information about AIBO such as the mass, center of mass, and inertias of the links of AIBO. Can I get this data? A. Sony is disclosing some part of AIBO (especially the software layer), but we still keep the hardware layer hidden. The physical characteristics are an extremely important part of a robot's design. Please understand that some parts of the internals cannot be disclosed because of commercial business reasons. |
| V 1662 gcc update? ejt [2005/04/09-03:40] |
| Response : HomePage : Email : Edit : Delete |
|
I
was recently playing with the precompiled header support of the gcc 3.4
series on my desktop machine, and was able to reduce compile time by 25-40%, depending on how many files needed to be recompiled and how broad I made the common header. This is a pretty significant savings, and further, I have some anecdotal evidence that 3.4 is itself a bit faster than 3.3 (the current version) even without PCH. The 3.4 branch has had time for several bug fix releases and generally seems quite stable, so I'd say it's mature enough to switch over (but you will probably want to research this a bit more thoroughly than I have before doing anything official :) So, it would save a lot of people a lot of time if someone took it upon themselves to update the build- devtools scripts to use 3.4... volunteers? :) (looking in Sony's general direction...) (PS I'd still like to see a response to https://openr.aibo.com/cgi-bin/openr/e_regi/im_trbbs.cgi? uid=general&df=bbs.dat&prm=TAN&pg=1&no=1638#1638 ) |
| V 1658 Help with Client App.. laue [2005/04/01-10:04] |
| Response : Email : Edit : Delete |
|
Hello, I'm working on the EchoClient app and I've changed the IP address to point to an internet site (google). I found google's IP and I changed the port to 80. But every time I try to receive a message, I keep getting back recvData : UNKNOWN 408 Request Timeout Server: thttpd Content-Type: text/html; charset=iso-8859-1 Date: Fri, 01 Apr 2005 01:00:35 GMT Last-Modified: Fri, 01 Apr 2005 01:00:35 GMT Accept-Ranges: bytes Connection: close Cache-Control: no- What else am I missing? I just simply want the Aibo client to hit a website in cyber space. Any help is appreciated. Thanks. |
| V 1656 wlan ad hoc oddness djtommyc [2005/03/22-06:16] |
| | Response : Edit : Delete |
|
| I'm
having an odd problem with the aibo and setting up an ad hoc connection. | When the Aibo is booting up the aibo and computer connect. However, as soon as | it( the aibo) is done booting up the as hoc connection is gone and never | resestablished. any ideas? | |
| +- 1657 RE:wlan ad hoc oddness AiboPet [2005/03/23-01:00] |
| Response : HomePage : Edit : Delete |
|
>I'm
having an odd problem with the aibo and setting up an ad hoc connection. >When the Aibo is booting up the aibo and computer connect. However, as soon as >it( the aibo) is done booting up the as hoc connection is gone and never >resestablished. any ideas? AIBO will only check for a WiFi connection once at boot time. You should have your PC up and running before turning on your AIBO. If the connection is dropped, you need to reboot the dog. Sounds like you have a marginal PC WiFi adapter (or perhaps lots of radio interference) I strongly suggest getting an access point for these configurations. This makes boot order less critical (ie. when the dog boots, it always connects to something) and reduces the chance of dropped connections greatly. NOTE: You don't need the access point attached to anything else (used just for the radio management) |
| V 1650 W3Aibo Sample explained... laue [2005/03/17-06:15] |
| | Response : Email : Edit : Delete |
|
| Hello, | | Is there any additional documentation for the w3Aibo sample program for Open-R? | | Thanks. | |
| +- 1652 RE:W3Aibo Sample explained... TodboT [2005/03/18-06:57] |
| | Response : Edit : Delete |
|
| If
you can read Japanese, there's an OPEN-R book that goes through the
sample | programs. There was supposed to be a book in English, but I don't think it has | been published yet. | |
| +- 1653 RE:W3Aibo Sample explained... laue [2005/03/19-03:09] |
| | Response : Email : Edit : Delete |
|
| Really?
I don't know Japanese. | | That's too bad. Do you think Google can translate it or some other program? | | >If you can read Japanese, there's an OPEN-R book that goes through the sample | >programs. There was supposed to be a book in English, but I don't think it has | >been published yet. | |
| +- 1655 RE:W3Aibo Sample explained... AiboPet [2005/03/20-01:15] |
| Response : HomePage : Edit : Delete |
|
>
Is there any additional documentation for the w3Aibo sample program for
Open- R? See my reply here: http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi? ubb=get_topic;f=5;t=001235 > ...Do you think Google can translate it or some other program? The book is available in print only (not in an online form that can be run through a program). There are some other The English version has been in the working for some time now (if and when they publish it, I assume it will be in print only) --- If you enjoy reading Google translated Japanese, you can search the Japanese portion of this website BBS. There are a few things not available to the English audience (a few pre-release RFW tutorials, some additional OpenR SDK topics etc). |
| V 1654 Broadcasting messages unconnected [2005/03/19-05:14] |
| Response : Email : Edit : Delete |
|
I'm
really stuck on this one. i can set up a udp server which bind to a specific port, and send data over it (using the udpecho-example) but now i want to let aibo broadcast a message. i just using the ip 192.168.1.255 to send to... but it fails. i guess i need to set up a 'broadcast' settings somewhere. can some-one help me out? |
| V 1647 Automation using Remote Framework... laue [2005/03/16-06:22] |
| | Response : Email : Edit : Delete |
|
| Hello, | | I have a question about using the remote framework. I want to develop a | program that allows Aibo to surf the internet. But I've been running into a | few snags so I'm contemplating on using the remote framework. | I have a concern though, in that I still want Aibo to be autonomous. I know | that the Aibo Entertainment Player has 2 modes (Remote/Auto), is that | programmable using the remote framework? How about Open-R? | | Thanks! | |
| +- 1648 RE:Automation using Remote Framework... TodboT [2005/03/16-07:32] |
| | Response : Edit : Delete |
|
| Aibo's
mode [auto/remote] is programmable. You should be able to create a | remote framework program that will connect to the Internet and get whatever | information that you're looking for. Once retrieved, you can have your PC | program change Aibo's mode from auto to remote, and then have AIBO react to the | information it retrieved. When done, simple switch the mode back to auto. | |
| +- 1649 RE:Automation using Remote Framework... laue [2005/03/17-00:34] |
| | Response : Email : Edit : Delete |
|
| Thanks
for the advice. Do you know if there are any other Remote Framework | tutorials available besides what is on this website? | | Thanks, | | >Aibo's mode [auto/remote] is programmable. You should be able to create a | >remote framework program that will connect to the Internet and get whatever | >information that you're looking for. Once retrieved, you can have your PC | >program change Aibo's mode from auto to remote, and then have AIBO react to | the | >information it retrieved. When done, simple switch the mode back to auto. | |
| +- 1651 RE:Automation using Remote Framework... TodboT [2005/03/18-06:51] |
| Response : Edit : Delete |
|
Sorry,
I'm not aware of any other Remote Framework guides. |
| V 1645 AIBO PlayMotion: CommandList.txt randeep [2005/03/15-20:57] |
| | Response : Edit : Delete |
|
| Hi, | | I could not find CommandList.txt file, which has the motion details that can be | used in the following function: | | Aibo->PlayMotion(int mtn); | | I could only find for Hello, Bye and Dance from the TakePicture example in RFW. | | for Hello mtn = (0x11 << 24) | (0x00 << 16) | (0x00 << 8) | 0x05 | for Bye mtn = (0x11 << 24) | (0x00 << 16) | (0x03 << 8) | 0x00 | for Dance mtn = (0x15 << 24) | (0x03 << 16) | (0x00 << 8) | 0x00 | | Please let me know what are the other values for Aibo motion. | | Thanks, | Randeep | |
| +- 1646 RE:AIBO PlayMotion: CommandList.txt AiboPet [2005/03/16-00:59] |
| Response : HomePage : Edit : Delete |
|
>I
could not find CommandList.txt file, which has the motion details that
can be >used in the following function: > >Aibo->PlayMotion(int mtn); BACKGROUND: The 'PlayMotion' function can be used in two different ways. First you can pass in an LMS value (32 bit value, L<<24|M<<16|S<<8|level). This is a higher level abstaction of a motion. It may perform different skits depending on the dog's current posture. Another way is to pass in a MWCID (16 bit number). This gives you explicit control over the dog (ie. you must specify which Hello skit you want -- whether the sitting version, the standing version, or whatever) ----- Take a look at "RFW_CDB_E_Release1.XLS" in the RFW SDK. It is an Excel spreadsheet (.csv version also included) This lists the motions available in AiboMind2/RFW with both their LMS+level values as well as the MWCID. Notice there are several rows that have the same L/M/S values (but different MWCIDs) NOTE: there are 3099 of them listed. They are the most useful (over 5000 total in the ERS-7.MWC file used by the AiboMind2 personality) ------- NOTE: the LMS/MWCID codes for RCODE ERS-7 are different. They are documented in the command-ers7-*.txt file in the RCODE SDK. |
| V 1643 Email problems laue [2005/03/12-07:51] |
| | Response : Email : Edit : Delete |
|
| Hello all, | | I'm new to the Aibo and have only been playing with him for about a month. | Here is my problem: | | I am able to connect to the Aibo from the WLanManager fine. I can connect to | the AEP with no problems. However, when I try to set up email, the connection | check pages always tell me I failed. I've set up my email accounts on outlook | with no problems. | | 1) Is the Aibo actually connecting to the network? | 2) Is there any way I can check the connection status ON the Aibo | 3) Why are the Aibo's ears only blue when the AEP is running? | 4) Are there any good FREE pop/smtp email providers available? | | I'm not sure what the problem is. I've tried almost everything. Please | help!!! | | Technical particulars: | | # | # WLAN | # | HOSTNAME=AIBO | ESSID=CPSC | WEPENABLE=0 | #WEPKEY=AIBO2 | APMODE=1 | | # | # IP network | # | USE_DHCP=1 | | # | # SSDP | # | SSDP_ENABLE=1 | | Email used: | | pop: mail.bluebottle.com | smtp: smtp.bluebottle.com | email addr: iolaus(AT)bluebottle.com | email account: iolaus | pw: ************ | | Thanks! | |
| +- 1644 RE:Email problems AiboPet [2005/03/13-01:05] |
| Response : HomePage : Edit : Delete |
|
>I
am able to connect to the Aibo from the WLanManager fine. I can connect
to >the AEP with no problems. That is a good first step. >However, when I try to set up email, the connection >check pages always tell me I failed. I've set up my email accounts on outlook >with no problems. This is much more difficult. >1) Is the Aibo actually connecting to the network? Yes, it is trying to go directly to the Internet, looking up your email accounts and extracting email messages. In your case I assume you have a WiFi Access Point connected directly to the Internet (cable modem or DSL). The DHCP feature of the access point should be providing the necessary connection info (DNS server, reachable IP address etc) This makes it much harder to setup. >2) Is there any way I can check the connection status ON the Aibo Try the AIBO webpage's "Connection Check". http://10.0.1.100 (or whatever IP address your Access Point allocates to AIBO) ----- In general I suggest asking AIBO setup and user questions on the public aibo- life.org BBS. There are quite a few LAN setup FAQs, as well as many more people reading the BBS (this BBS is better for OpenR and other "SDE" questions) http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi ----- FWIW/IMHO: the AiboEyes/AiboMail feature is not worth all the setup hassle. |
| V 1639 timers sf [2005/03/05-04:53] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi. | | I am new to OpenR programming. It claims to be a fully object-oriented | distributed real-time system. However, after reading some of the documentation, | I have some questions that you might be able to answer: | | - Are there any timers available? | | - Is there a way to access the scheduling strategy? | | - Is it possible to replace OpenR with a different chip-compatible operating system? | | - Can we access interrupts? | | Thanks a lot, | Sebastian | |
| +- 1640 RE:timers AiboPet [2005/03/06-01:48] |
| Response : HomePage : Edit : Delete |
|
>
I am new to OpenR programming. It claims to be a fully object-oriented >distributed real-time system. First question: where did you read such a claim? Although technically true, it is object-oriented (at a very simple level), supports distribution (in a limited fashion using Remote OpenR) and certainly provides real-time system control for things that need it. Then again so does Windows ;-> > However, after reading some of the documentation, >I have some questions that you might be able to answer: > >- Are there any timers available? Not directly. There are timing tricks [see Tekkotsu for a framework that packages up common features like timers, done the Tekkotsu way - not in raw OpenR] >- Is there a way to access the scheduling strategy? No. There are fixed module priorities. >- Is it possible to replace OpenR with a different chip-compatible operating system? No. The interface below the official OpenR SDK is not documented (and most of the core code is encrypted to prevent you from replacing it too) >- Can we access interrupts? No. This is somewhat moot since you don't know how the interrupts work. ===== NOTE: all the above answers are "officially no". There is a lot of internal details you can figure out by looking at the headers for parts that aren't in the officially documented SDK. When you get too low level, you hit encrypted code modules and the undocumented hardware itself. Bottom line: it goes back to expectations. When you read the marketing claims ("object oriented, distributed, realtime") don't read too much "plugability" or "replacability" into those features. |
| V 1633 Display of AIBO 3D object in Directx Fairlawn07 [2005/02/24-05:29] |
| | Response : Email : Edit : Delete |
|
| Hi, | I am trying to display AIBO 3D object in Directx program. When I try to use | CAIBO3D::Create3D I will see the image only for short time and then it will | disappear. Using D3DXLoadMeshFromX will display the AIBO 3D, but then I have | an issue how to connect the sensors from "REAL AIBO" to the AIBO 3D. Any | suggestions or code snippets are appreciated. | Regards, Henry | |
| +- 1635 RE:Display of AIBO 3D object in Directx TodboT [2005/03/03-10:40] |
| | Response : Edit : Delete |
|
| I
haven't really looked, but what you're looking for is probably in the | RemoteTest sample. | |
| +- 1637 RE:Display of AIBO 3D object in Directx Fairlawn07 [2005/03/03-12:44] |
| Response : Email : Edit : Delete |
|
>I
haven't really looked, but what you're looking for is probably in the >RemoteTest sample. Thanks, but RemoteTest sample is for windows application and not for Direct X. In vaibo7_spec.html specifications it states: "A 3D AIBO model can be displayed using Direct X". Regards, Henry |
| V 1632 R-CODE for ERS-7 (ver.2) has been released openrsupport@(Administrator) [2005/02/22-08:05] |
| Response : Edit : Delete |
|
Hello, R-CODE for ERS-7 (ver.2) has been released, and is available for download now. * A bug in R-CODE's numeric processing has been fixed. * 32bit intergers are now treated correctly. * RRLeg (right rear) sensor data can now be read correctly. Thanks, AIBO SDE Support |
| V 1628 setting up ad hoc djtommyc [2005/02/17-12:50] |
| | Response : Email : Edit : Delete |
|
| I've
been working on setuping up my Aibo to work with my new laptop (I had it | working just peachy with the old one) and I've run into a problem. When | setting up the adhoc network on the new laptop, I now no longer have the option | to specify the gateway, IP address, channel, etc. There problem seems to lay in | the software for the onboard wireless card. Does anyone know of 3rd party | drivers that may give me more options when setting up an adhoc network or a site | that is a good source of information for this type of thing? | | Thanks for the help | |
| +- 1631 RE:setting up ad hoc AiboPet [2005/02/19-03:05] |
| Response : Edit : Delete |
|
>
When setting up the adhoc network on the new laptop, I now no longer
have the option to specify the gateway, IP address, channel, etc Find the drivers that came with the card or search on the web for newer ones. Setting up adhoc *requires* the ability to set the IP address [or at least have a private IP picked for you] |
| V 1629 led face Iaia [2005/02/17-16:59] |
| | Response : Email : Edit : Delete |
|
| Hi! | Can I have .led files? How can I make .led files to myself? | |
| +- 1630 RE:led face AiboPet [2005/02/19-03:03] |
| Response : Edit : Delete |
|
>Hi! >Can I have .led files? How can I make .led files to myself? .LED files are useful for RCODE personalities YART_APG/YART_APG7 will automatically generate .LED files for your sound files (the Eyes will flash in time to the audio) http://www.aibohack.com/ers7/yart7.htm (ERS-7 and ERS-2x0 are different versions) Skitter can be used for more manual editing of .LED files http://www.dogsbodynet.com For normal use, the .LED files must be combined into an .ODA file (LED.ODA) and used with RCODE functionality. There are similar options for use with Mind2/RFW (create .ODA and upload to dog before trying to play) |
| V 1626 OPEN-R on CYGWIN sshpark [2005/02/15-04:51] |
| | Response : Email : Edit : Delete |
|
| Hello, | I've been trying to compile a sample program using CYGWIN and Windows XP. I | have installed the correct version of cygwin,mipsel-devtools and OPEN_R_SDK as | it was displayed in the download page. However, when I compile the sample | program, and boot up AIBO, absolutely nothing happens. I can't even turn it | off. Did anyone encounter this problem before? | Thank you in advance. | | -Shelly | |
| +- 1627 RE:OPEN-R on CYGWIN AiboPet [2005/02/17-02:13] |
| Response : Edit : Delete |
|
If
you have an old ERS-210 model (not-A) then you may need to reflash the system ROM (see the downloads page). Not a problem with later 210s, or any 210A, 220, 220A or ERS-7s Here are some common install problems that people encounter: + Not following the installation instructions completely + Not using a PMS/RCODE stick http://www.aibohack.com/pms.htm + Not installing the proper system files to memory stick before installing the program + If installing the WCONSOLE version, not having a telnet session for debugging + Typing 'make' instead of 'make install' + Forgetting to copy the results of 'make install' to the stick + Starting with 'HelloWorld' (it doesn't do anything except print to the console). Try "BallTrackingHead" or other sample instead |
| V 1625 Author DELETE ***** [2005/02/15-04:17] |
| . |
|
(The
article that 2005/02/15-04:07 posted, was modified by Author) |
| V 1619 compiling errors for remote processing robert25 [2005/02/03-14:27] |
| | Response : Edit : Delete |
|
| Hi: | I was involved in a uni project that needs to program on the AIBO. | I ran the "setup-rp-openr" script all right. But when I try to run "make | install" in the sample/common/ObjectComm directory, I get error message " | ***no rule to make target 'and'.Stop. | | Tried it many times even reinstalled the whole thing. | Your advice in this matter is highly appreciated. | | Robert | |
| +- 1624 RE:compiling errors for remote processing TodboT [2005/02/15-02:30] |
| Response : Edit : Delete |
|
>***no
rule to make target 'and'.Stop. It looks like you typed "make and" instead of "make install". |
| V 1622 Low-level Wireless bowling_m_h [2005/02/05-15:33] |
| Response : HomePage : Email : Edit : Delete |
|
I
have been looking into doing localization on the AIBO using the wireless signal as a sensor. Looking at the headers for WLAN information I don't see any way to find out, which access point (MAC Address?) the AIBO is connected to. Is this not accessible? Ideally, I could get a list of available access points and their signal strengths to be used to triangulate. Is there some hidden functionality that might make this possible. Thanks for any help. Michael |
| V 1615 AIBO's sensorimotor system bpaul [2005/02/02-22:17] |
| | Response : HomePage : Edit : Delete |
|
| Hello all, | | Quote: "The first generation of AIBO had acceleration sensors in each leg, | angular velocity sensors also in each leg, touch sensors in the head and each | paw and two heat sensors in the torso. The acceleration sensors together with | the angular velocity sensors enhanced movements. They were used, for example, to | maintain balance, for regaining balance when recovering from a fall and for | detecting rotary motions." | | How are the 3(x,y,z)acceleration sensors of ERS-7 different from the one's in | the first generation ? | | Has anyone of you ever analysed how the accelerometer readings change with the | motion of the leg joints and how that further helps determining the spatial | organization of the legs ? | |
| +- 1617 RE:AIBO's sensorimotor system AiboPet [2005/02/03-01:52] |
| | Response : Edit : Delete |
|
| Commentary/speculation: | | >Quote: "The first generation of AIBO had acceleration sensors in each leg, | >angular velocity sensors also in each leg... | Not sure where that quote came from. | | My guess is they are referring to the legs measuring and calculate | acceleration and angular velocity [ie. part of the PID algorithm], not | separate accelerometers on each leg (that would be a lot) | | IIRC the ERS-11x has the same accelerometers in the body like all the models | (ERS-2x0 and ERS-7 have all 3, but the ERS-31x only has 2). They are at | minimum used for tilt detection (fall down) and fast liftup detection in the | commercial aiboware. | | Unclear what else is used in the commercial AiboWare since that is Sony | proprietary. There is room in the walking models (including the .MTN files) | for the orientation of the body (X,Y,Z angles). I don't know if it was / is | used. | | Anecdotal: Usually AIBO will continue to do things exactly the same when you | pick him up, and tilt him slightly (if you tilt him too far, he will kick into | falldown recovery mode) | |
| +- 1620 RE:AIBO's sensorimotor system bpaul [2005/02/03-21:44] |
| | Response : Edit : Delete |
|
| According
to the readings i've seen the displayed variations of the acceleration | sensors during forward/backward-left/right motion are all ambigious and there is | no definite conclusion that one can draw from them, except that the sensors are | unrelaible and therefore can not be used. | > | >My guess is they are referring to the legs measuring and calculate | >acceleration and angular velocity [ie. part of the PID algorithm], not | >separate accelerometers on each leg (that would be a lot) | > | Are there any more details on the PID algorithm being used by Aibo's legs ? | > | >IIRC the ERS-11x has the same accelerometers in the body like all the models | >(ERS-2x0 and ERS-7 have all 3, but the ERS-31x only has 2). They are at | >minimum used for tilt detection (fall down) and fast liftup detection in the | >commercial aiboware. | > | Yes, I guess for that the accelZ senosors are the ones being used for that(the | only 1 sensor out of the total 3 whose readings make some sense) | > | >Unclear what else is used in the commercial AiboWare since that is Sony | >proprietary. There is room in the walking models (including the .MTN files) | >for the orientation of the body (X,Y,Z angles).I don't know if it was/is | >used. | > | Probably yes, there are some soccer-teams that are using them for the same | purpose, but I have no idea how ? | > | >Anecdotal: Usually AIBO will continue to do things exactly the same when you | >pick him up, and tilt him slightly (if you tilt him too far, he will kick into | >falldown recovery mode) | |
| +- 1621 RE:AIBO's sensorimotor system ejt [2005/02/04-02:16] |
| Response : HomePage : Email : Edit : Delete |
|
>Are
there any more details on the PID algorithm being used by Aibo's legs ? We did a little experimentation, so you can convert between a regular single floating point value for each of P,I,D and the gain/shift arguments which are required by the system for each. http://www-2.cs.cmu.edu/%7Etekkotsu/dox/classPIDMC.html There's a lot of literature on PID control in general, although there are a couple variations. I think they are using the most common form where each parameter contributes independently, but I could be wrong. >Probably yes, there are some soccer-teams that are using them for the same >purpose, but I have no idea how ? If you do some statistical analysis of the acelerometers, you can tell the difference between normal walking and walking into an obstacle (e.g. caught on another player or out-of-bounds barrier) |
| V 1618 Speaker Problem -> Broken Robot Sauber [2005/02/03-14:24] |
| Response : Email : Edit : Delete |
|
Hi, It appears we have broken ERS-7 robot. The error (with no memory stick) is shown below. The problem looks to occur when first contacting the speaker (in our code the robot turns off during our sound module initialisation). The odd thing is that the robot still plays the 'crash' sounds; therefore the speaker is working in some manner. It should be noted that in our code we simply get the PRM speaker error not the CMAgent errors that occur during the no memory stick boot. So it is an error relating to the speaker that shuts down the robot. We are wondering two things: 1. Has anybody seen this particular error before? Is it really the speaker or is the speaker just the first sign of a system wide problem? 2. If the problem is the speaker, is there a way to force the robot to not load the speaker? The boot routine appears to contact the speaker even if you never use the speaker in your code. It seems a bit impractical that the robot shuts down if only the speaker is broken. One of the initial aims of OPEN-R was to make a library that can adapt to different configurations of a robot, since the speaker is a non-critical component it should boot even if no speaker was present. Thanks, Michael (NUbots) Error ------ [oid:0x8000004f] /FL/OPEN-R/APP/OBJS/IC.BIN [oid:80000042,prio:1] Prelude::Can't OpenPrimitive(speaker) [oid:80000049,prio:1] CMAgent[1100]::oprimitiveID Error!!! 0:PRM:/r1/c1-Joint2 :11 6 [oid:80000043,prio:1] CMAgent[1130]::oprimitiveID Error!!! 0:PRM:/r1/c1-Joint2 :11 6 [oid:80000043,prio:1] [oid:80000045,prio:1] CMAgent[1140]::oprimitiveID Error!!! 0:PRM:/r4/c1-Joint2 :41 6 [oid:80000047,prio:1] CMAgent[1010]::oprimitiveID Error!!! 0:PRM:/r1/c1-Joint2 :11 6 [oid:80000046,prio:1] CMAgent[3000]::oprimitiveID Error!!! 0:PRM:/r1/c1/c2/c3/ l3-LED2:l3 6 [oid:8000004a,prio:1] SP::Can't Open Speaker PRM [oid:8000004b,prio:1] CM: Can't Open primitiveID [oid:80000045,prio:1] LG::SetPrimitive Error [oid:8000004b,prio:1] CM: Montella will not play [oid:80000045,prio:1] OPENR::Fatal(0) [oid:80000047,prio:1] CMAgent[1020]::oprimitiveID Error!!! 0:PRM:/r1/c1/c2/c3/ c4-Joint2:14 6 [oid:80000046,prio:1] CMAgent[3003]::oprimitiveID Error!!! 0:PRM:/r1/c1/c2/c3/ l6-LED2:l6 6 [oid:8000003c,prio:1] OSystemLogger::Fatal() : powerManager_.WriteDataStream() F AILED 2 OPowerManager::PowerOff() SHUTDOWN ... |
| V 1610 AIBO's conversational skills andreshs1 [2005/02/01-05:58] |
| | Response : Email : Edit : Delete |
|
| Hi there, I am very new to AIBO's culture. | | I am a student researching into AIBO's capabilities for cohabiting with | elderly people in order to help them out carrying random tasks around the | house. | Initially the problem I find is the limitation on memory, how can you overcome | the problem when creating multiple applications to be run on AIBO | simultaneously? | | Is AIBO able to synthesize sound, like a voice in order to maintain | conversations? | | Are you aware of any work on this area? | | Thanks | |
| +- 1613 RE:AIBO's conversational skills TodboT [2005/02/02-02:28] |
| | Response : Edit : Delete |
|
| I
don't know of any text-to-speech or voice recognition in OPEN-R, but
you could | use Remote Framework. However, Aibo will be tethered to a PC. Remote Framework | has a TTS feature and you can use the limited voice recognition from the Mind | software. You can also grab the audio and feed it into a different voice | recognition engine on your PC. I believe AiboPet has already done something | like this. | | TodboT | |
| +- 1616 RE:AIBO's conversational skills AiboPet [2005/02/03-01:41] |
| Response : Edit : Delete |
|
>
I believe AiboPet has already done something >like this. BrainBo (older version) and HomieGate (newer version) http://www.aibohack.com/homie http://www.aibohack.com/210/brainbo.htm Demo movie: http://www.aibohack.com/movies/apv_brainbo_lores.wmv ---- A future HomieGate Beta 5 will support RFW in AiboMind2 (whenever I bother to clean it up and release it) |
| V 1614 i lost my AIBO mind Japanese version lisans [2005/02/02-19:51] |
| Response : Email : Edit : Delete |
|
i'm change my AIBO mind to English today and i forget backup my Japanese mind before upgrade any one know how to get my Japanese AIBO mind backup ? please help me and thank you very much. |
| V 1609 aibo dead nlopes [2005/02/01-02:18] |
| | Response : Email : Edit : Delete |
|
| Hi, | | Today I was playing around with one aibo ERS-7 & Teckkotsu and suddenly the aibo | died! | Now it doesn't even switch on. I've already tried other batteries and other | memory sticks (just in case), but it is still dead. (the power light is off). | | Our aibos were recently inspected because they were in "sony's blacklist". (they | have replaced some parts). | | So, now what should I do? Any suggestions? | | Thanks, | Nuno | |
| +- 1612 RE:aibo dead TodboT [2005/02/02-02:18] |
| Response : Edit : Delete |
|
Nuno, Does the light turn orange if it is charging on the station? Does the station detect Aibo? Can you turn Aibo on if it's on the station? You should probably call Aibo Customer Link (1-800-427-2988) if you're still having problems. TodboT |
| V 1611 Announcing Tekkotsu 2.3 ejt [2005/02/01-06:25] |
| Response : HomePage : Email : Edit : Delete |
|
Tekkotsu 2.3 has been released! http://www.tekkotsu.org/ Notable new features of 2.3 include: * Ability to speak/flash the IP address (useful in unknown or dynamic networks) * VisionObjectEvents include bounding box information * Sensor Observer gives real time feedback within the Controller * Network signal strength status (thanks to Sony for ERA201D1 sample code) * new template files for state nodes, state machines, transitions, and controls More information is available from the Version History page: http://www-2.cs.cmu.edu/%7Etekkotsu/VersionHistory.html#v2.3 Enjoy! |
| V 1564 Cross-compiling libraries kolargol00 [2004/12/22-22:24] |
| | Response : Edit : Delete |
|
| Hello! | | I'd like to use the GNU Scientific Library (GSL -> | http://www.gnu.org/software/gsl/) in my Aibo control program. Does anyone know | how to cross-compile it for the Aibo's processor using the OPEN-R SDK? | |
| +- 1566 RE:Cross-compiling libraries ejt [2004/12/23-01:02] |
| | Response : HomePage : Email : Edit : Delete |
|
| >I'd
like to use the GNU Scientific Library (GSL -> | >http://www.gnu.org/software/gsl/) in my Aibo control program. Does anyone know | >how to cross-compile it for the Aibo's processor using the OPEN-R SDK? | | Usually it's a pretty straightforward matter of switching the compiler tools | that are normally used in a given library's Makefile to instead use the | mipsel-linux-* tools in OPEN_R_SDK/bin. | If it saves you some trouble though, Tekkotsu includes a copy of the newmat library: | http://www.robertnz.net/nm_intro.htm | newmat has a fair share of included numerical algorithms, although GSL looks | like it might have a few more... hmmm, I might look into that myself ;) | |
| +- 1605 RE:Cross-compiling libraries kolargol00 [2005/01/28-23:14] |
| | Response : Edit : Delete |
|
| >Usually
it's a pretty straightforward matter of switching the compiler tools | >that are normally used in a given library's Makefile to instead use the | >mipsel-linux-* tools in OPEN_R_SDK/bin. | | I tried that but the C compiler is not able to create executables (according to | configure script output). It turns out that compilation works with -c flag but | linking fails. | |
| +- 1606 RE:Cross-compiling libraries ejt [2005/01/29-04:00] |
| Response : HomePage : Email : Edit : Delete |
|
>I
tried that but the C compiler is not able to create executables
(according to >configure script output). It turns out that compilation works with -c flag but >linking fails. Yeah, the AIBO executables aren't quite the same as linux executables, for instance there's no main(). You can use mipsel-linux-ld to link things, for instance a library to be linked with other code, but when it comes time for the final executable, there is a special tool, 'mkbin' in 'OPEN_R_SDK/OPEN_R/bin'. I suggest looking at the Makefiles used in the sample projects if you're trying to compile an executable from scratch. |
| V 1604 Setting wireless interactively elucidatek [2005/01/27-16:25] |
| Response : Email : Edit : Delete |
|
I
would like to be able to scan for access points and select a specific
one and then be able to change to another access point as required. I know how to read the wlan information but I cannot figure out how to talk to the hardware. Does anyone have any suggestions or pointers. Has this been tried before? |
| V 1589 Determining AIBO's IP Address? ejt [2005/01/20-06:18] |
| | Response : HomePage : Email : Edit : Delete |
|
| Consider
doing a demo using a wireless network running DHCP, which you don't | have control over. If you hardcode your own address, it's a gamble whether | whatever address you choose is going to conflict with a previously assigned | address. But if you let DHCP assign an address, how do you know what address to | connect to? | | I'd like to write a little behavior which will flash the assigned IP address | through the LEDs on boot, which would solve this problem. (And it's a common | problem - many universities have their own wireless networks, so they don't like | it when users set up their own, and it can be a real nuisance to convince the | admins to figure out how to give your robot a static IP address) | | So, the question is, how do I get the current IP address? Several functions | will return the local ip address on the callback parameters, but it seems they | all only trigger the callback when a connection has been made, which isn't going | to happen before we know what address to connect to. Certainly there must be a | way to do this without requiring a connection to be made? | |
| |- 1590 RE:Determining AIBO's IP Address? AiboPet [2005/01/21-02:44] |
| | | Response : HomePage : Edit : Delete |
|
| | >
... So, the question is, how do I get the current IP address? | | | | Of course if you have control of the DHCP server you can know the IP addressed | | assigned. | | | | For the more general case, you need to use UPnP / SSDP | | | | ---- | | SSDP is built into RCODE and AiboMind software (\OPEN-R\APP\OBJS\SSDP.BIN) | | | | "SSDP_ENABLE=1" must be on in the WLANCONF.TXT | | Also the PC must have SSDP/UPnP enabled (disabled on older computers because | | of a historical security problem) | | | | The RFW SDK has a helper function to find AIBO for you. The WLAN Manager has a | | simple feature ("AIBO Search"). I think the helper will work with RCODE too. | | | | --- | | For Open-R SDK users, you must write your own SSDP server to run on the AIBO ;- | | < | | | | [or something equivalent that will respond to special broadcast events and | | send the PC the assigned IP address] | | |
| | +- 1591 RE:Determining AIBO's IP Address? ejt [2005/01/21-05:41] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | >>
... So, the question is, how do I get the current IP address? | | > | | >Of course if you have control of the DHCP server you can know the IP addressed | | >assigned. | | No, unfortunately the point is that when demoing at a remote location, or | | dealing with overworked/undertrained network administrators, we don't have | | control of the DHCP server. And I would postulate this is the case for a large | | segment of the market. (Even if the user does technically have control of DHCP, | | it's a pain to dig up the router's password, log in and try and figure out which | | address has been assigned, or set up a static one) | | | | So it would be very useful to have the dog report its IP directly, instead of | | trying to rely on knowledge of, permission, and access to external equipment. | | | | >For Open-R SDK users, you must write your own SSDP server to run on the AIBO ;- | | Which brings me back to the original question... If I were to do something like | | this, something, somewhere, must know the IP address before we can broadcast it. | | I need access to that something before I can do anything. :( | | |
| | +- 1593 RE:Determining AIBO's IP Address? AiboPet [2005/01/22-01:39] |
| | | Response : HomePage : Edit : Delete |
|
| | ?
am I missing something here ? | | | | Are you saying you can setup a server process on AIBO and still not know | | AIBO's IP address ? | | | | I don't use the DHCP approach (ie. in the non-DHCP case, it will get the IP | | address from WLANCONF.TXT when needed, in this case when you initialize the | | WiFi for setting up the server processes) | | | | ------ | | >>For Open-R SDK users, you must write your own SSDP server to run on the | | AIBO ;- | | | | Or something like SSDP which starts with a general broadcast from the PC that | | AIBO picks up. | | The AIBO must set up a server process to listen for the broadcast. AIBO should | | have an IP address by the time it is running. | | |
| | +- 1594 RE:Determining AIBO's IP Address? ejt [2005/01/22-01:55] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Are
you saying you can setup a server process on AIBO and still not know | >AIBO's IP address ? | You can tell the network stack to listen for a connection, but I haven't found a | way to simply ask the network stack what the IP is. I'm sure the network stack | knows the IP address (it has to), but getting that information myself is the | problem. The only way I can find to get the address is after a connection is | made, and the network stack is reporting that fact to the callback function. | | If there's some function which returns the current IP address, that's all I'm | asking, you're completely right, this *shouldn't* be a difficult thing. I was | hoping there was something obvious I'm missing. | | >Or something like SSDP which starts with a general broadcast from the PC that | >AIBO picks up. | >The AIBO must set up a server process to listen for the broadcast. AIBO should | >have an IP address by the time it is running. | This is a pretty good idea, in fact it may be the only way to go, unless someone | can simply tell me how to determine the IP address *before a connection is made*. | I haven't done much with packet broadcasting, so I was really hoping to just do | something simple with blinking a number on the LEDs. | |
| +- 1600 RE:Determining AIBO's IP Address? openrsupport@(Administrator) [2005/01/26-12:19] |
| | Response : Edit : Delete |
|
| Please
see the ERA201D1Info sample program. This sample will show you how to | obtain information about the Wireless LAN settings in AIBO. You can read the | MAC address, Ethernet statistics, WLAN settings, WLAN statistics, and the IP | address. | | Thank you, | | AIBO SDE Support | |
| |- 1601 RE:Determining AIBO's IP Address? TodboT [2005/01/26-12:23] |
| | | Response : Edit : Delete |
|
| | I
think it would be cool if you have aibo "say" its IP address back to
you. | | You'd only need eleven recordings. That would be easier to understand than | | trying to read the face LEDs. | | | | TodboT | | |
| | +- 1603 RE:Determining AIBO's IP Address? ejt [2005/01/27-01:25] |
| | Response : HomePage : Email : Edit : Delete |
|
| Dave
suggested the same thing -- Keep an eye out for 2.3 in the near future
;) | >I think it would be cool if you have aibo "say" its IP address back to you. | >You'd only need eleven recordings. That would be easier to understand than | >trying to read the face LEDs. | > | >TodboT | |
| +- 1602 RE:Determining AIBO's IP Address? ejt [2005/01/27-01:22] |
| Response : HomePage : Email : Edit : Delete |
|
Aha!
Thanks... I did miss something semi-obvious ;) I was looking in all the Print...() functions for the part where an IP is obtained, but this one item is simply directly included in the DoStart() which calls all of the Print...() functions, so I missed it. Thanks! -ethan PS For those following at home, the code in question is: antStackRef ipstackRef = antStackRef("IPStack"); antEnvInitGetParamMsg getParamMsg("ETHER_IP"); getParamMsg.Call(ipstackRef, sizeof(getParamMsg)); if (getParamMsg.error == ANT_SUCCESS && getParamMsg.paramType == antEnv_InitParam_String) { OSYSPRINT(("[ETHER_IP] %s\n", getParamMsg.value.str)); } else { OSYSLOG1((osyslogERROR, "getParamMsg.Call() FAILED %d", getParamMsg.error)); } (The article that 2005/01/27-01:20 posted, was modified by Author) |
| V 1569 receiving and sending buffers to be displayed? xph [2005/01/01-03:06] |
| | Response : Edit : Delete |
|
| Hi,
I was wondering if there is any way to add something to RFW (remote
test, | for example) that streams the image data from the robot onto buffers, makes | certain modifications to the buffer, then sends the buffer back into the code | to be displayed? It seems right now I can grab the buffer from the camera | image and modify it, but I cannot figure out how to send the modified buffer | back to the pointer (or whatever it is) to be displayed. Is there some | function that lets me do this? any help would be appreciated. thanks | |
| +- 1570 RE:receiving and sending buffers to be displayed? AiboPet [2005/01/03-02:14] |
| | Response : HomePage : Edit : Delete |
|
| >
(is there a way to stream) the image data from the robot onto buffers | | RemoteTest uses the direct display approach (using 'ImageSetWindowHandle'). | | Look at "ImageGetData". I can be called in the WM_IMAGETP_UPDATE notification | to get the image data, in JPG format. | Setting up for image capture requires some initialization. Look at | the 'TakePicture' sample | | After you have the image data in JPG format, you can decode it and display the | image, or do whatever you want with it. | |
| +- 1595 RE:receiving and sending buffers to be displayed? xph [2005/01/22-03:54] |
| | Response : Edit : Delete |
|
| Thanks a lot | | I was wondering if there is any way, after I manage to get image data (in the | form of a buffer) and make whatever modifications I see fit to the buffer, to | send the buffer back into the client to be outputted directly. As I understand | it, the RFW stores the image data in some pointer that is inaccessible to the | user (right?). It would be nice if there were a way to take my modified data | buffer and replace the original buffer stored in that hidden pointer, so the | client can automatically display my images instead of its original ones. But | is this possible? | | Also, I still seem to have trouble with ImageGetData. Every time I try to call | it, it gives me the following error: | | Unhandled exception at 0x003d5e29 in RemoteTest.exe: 0xC0000005: Access | violation reading location 0x019e0000. | | I've done everything I can possibly think of to fix it, but nothing works. Any | ideas? Thanks in advance. | |
| +- 1596 RE:receiving and sending buffers to be displayed? xph [2005/01/22-04:27] |
| | Response : Edit : Delete |
|
| >Thanks
a lot | > | >I was wondering if there is any way, after I manage to get image data (in the | >form of a buffer) and make whatever modifications I see fit to the buffer, to | >send the buffer back into the client to be outputted directly. As I understand | >it, the RFW stores the image data in some pointer that is inaccessible to the | >user (right?). It would be nice if there were a way to take my modified data | >buffer and replace the original buffer stored in that hidden pointer, so the | >client can automatically display my images instead of its original ones. But | >is this possible? | > | >Also, I still seem to have trouble with ImageGetData. Every time I try to call | >it, it gives me the following error: | > | >Unhandled exception at 0x003d5e29 in RemoteTest.exe: 0xC0000005: Access | >violation reading location 0x019e0000. | > | >I've done everything I can possibly think of to fix it, but nothing works. Any | >ideas? Thanks in advance. | | I guess I'll add a little more info to my last query, so I can get better and | more specific help. | | As for the WM_IMAGETP_UPDATE notification, the only thing I found on it in the | entire application is that it leads to the call of | CRemoteTestDlg:OnITPImageUpdate which returns OK and does nothing else. I tried | to call ImageGetData in this function as well but got the same error. Here is | how I called (maybe I'm doing something wrong syntactically?): | int bufLen = 97308; | unsigned char buf [97308]; | m_vaibo->ImageGetData(buf, bufLen); | | I've also tried variations, such as | int bufLen = 97308; | unsigned char *buf = new unsigned char; or = new unsigned char [bufLen]; | m_vaibo->ImageGetData(buf, bufLen); | | And same problem. What exactly am I doing wrong? Any ideas? Thanks, appreciate | it a lot. | |
| +- 1597 RE:receiving and sending buffers to be displayed? AiboPet [2005/01/22-07:27] |
| | Response : HomePage : Edit : Delete |
|
| >
I was wondering if there is any way, after I manage to get image data
(in the | > form of a buffer) and make whatever modifications I see fit to the buffer, to | > send the buffer back into the client to be outputted directly. | | Not AFAIK. | There is the direct-and-easy method (ImageSetWindowHandle and forget) and the | do-it-yourself more flexible method (ImageGetData and do the rest yourself) - | no middle ground. | ----- | | > What exactly am I doing wrong? | | You are ignoring my earlier advice ;-) | | Repeating and elaborating: | | Setting up for image capture requires some initialization. Look at | the 'TakePicture' sample. | | TakePicture does the initialization the correct way for ImageGetData (JPEG). | RemoteTest doesn't not. | | I have a simple command line capture program if you need it (grabs JPGs from | the AIBO), but it was adapted from the 'TakePicture' logic. | |
| +- 1598 RE:receiving and sending buffers to be displayed? xph [2005/01/24-14:28] |
| | Response : Edit : Delete |
|
| >TakePicture
does the initialization the correct way for ImageGetData (JPEG). | >RemoteTest doesn't not. | > | >I have a simple command line capture program if you need it (grabs JPGs from | >the AIBO), but it was adapted from the 'TakePicture' logic. | | Thanks a lot for your help! Can you send me this simple image collection | program? Just enough code allow me to make an ImageGetData call in order to | extract image data from the AIBO. Thanks again!! | |
| +- 1599 RE:receiving and sending buffers to be displayed? AiboPet [2005/01/25-01:55] |
| Response : HomePage : Edit : Delete |
|
>
Can you send me this simple image collection program? Sure, send me an email aibopet@aibohack.com |
| V 1530 Bug Report for IPSTACK object jcbaillie [2004/11/23-18:42] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi, | | You will find here: | http://www.ensta.fr/~baillie/bug_report.pdf | | a full bug report regarding a critical issue with the IPSTACK OPENR object. | | The only workaround we found was to fill the sending buffer with a size between | 1025 up to 3072 octets up to 3073 octets, thus avoiding the 1025-3072 range and | the problem. | | JC Baillie | contact: saltel at ensta.fr | |
| +- 1542 RE:Bug Report for IPSTACK object openrsupport@(Administrator) [2004/12/02-02:20] |
| | Response : Edit : Delete |
|
| Thank
you for the bug report about the IPSTACK OPEN-R object. | We are analyzing the problem but this is our 1st answer for you. | | 1) It is normal in a TCP/IP network that a packet is | sliced and the receiver gets two or more packets. | The application program should handle separated packets. | | 2) There is busy wait code in the SENDER side of the program. | // Wait for the Receiver object to treat the sent packet | (deb = mtime(), j = deb; j < deb + 2000; j = mtime()); | This code blocks the process running in AIBO. | Busy wait code can cause serious problems in OPEN-R programming. | But we cannot say that this code is the reason that the response | becomes so bad. | | 3) We are still checking if IPSTACK has a problem or not. | It will take some time to get this result. | IPSTACK is used in AIBO MIND2 and it is a core object for | network communication, especially for AIBO Remote Framework. | And we have tested IPSTACK in oppressive conditions. | | Thank you, | | AIBO SDE Support | |
| |- 1544 RE:Bug Report for IPSTACK object msgid [2004/12/03-03:04] |
| | Response : Edit : Delete |
|
| > | >1) It is normal in a TCP/IP network that a packet is | >sliced and the receiver gets two or more packets. | >The application program should handle separated packets. | > | It is true, but I'm implementing a comunication protocol message oriented over | TCP and I like to know if there's a way to force IPStack to send the collected | data without split the package. | |
| |- 1551 RE:Bug Report for IPSTACK object saltel [2004/12/15-23:26] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi, | | I work with JC Baillie and I'm responsible for the bug report | regarding the issue with the IPStack Object. Thank you for your answer. | | >1) It is normal in a TCP/IP network that a packet is | >sliced and the receiver gets two or more packets. | >The application program should handle separated packets. | | The issue isn't about the transmission of packets over the | TCP/IP network but it consists in finding the reason why ReceiveCont() | callback function in the RECEIVER side of the program is | called twice when sent packets' size stands between 1025 up to 3072 | octets but is called once in all other cases. | | >2) There is busy wait code in the SENDER side of the program. | >// Wait for the Receiver object to treat the sent packet | >(deb = mtime(), j = deb; j < deb + 2000; j = mtime()); | >This code blocks the process running in AIBO. | >Busy wait code can cause serious problems in OPEN-R programming. | >But we cannot say that this code is the reason that the response | >becomes so bad. | | I have changed the code in order not to block the running process anymore | and the issue remains. I have replaced the "for" expression | by the NotifyImage() function called every 32 milliseconds. | | You will find the updated source code and bug report at : | http://www.ensta.fr/~saltel | | Thank you for your support. | | B. Saltel | | contact : bastien.saltel@free.fr | |
| +- 1585 RE:Bug Report for IPSTACK object jcbaillie [2005/01/17-18:42] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi, | | Is there any news from the openr team regarding this bug report? | Thanks, | JCB | | >Thank you for the bug report about the IPSTACK OPEN-R object. | >We are analyzing the problem but this is our 1st answer for you. | > | >1) It is normal in a TCP/IP network that a packet is | >sliced and the receiver gets two or more packets. | >The application program should handle separated packets. | > | >2) There is busy wait code in the SENDER side of the program. | >// Wait for the Receiver object to treat the sent packet | >(deb = mtime(), j = deb; j < deb + 2000; j = mtime()); | >This code blocks the process running in AIBO. | >Busy wait code can cause serious problems in OPEN-R programming. | >But we cannot say that this code is the reason that the response | >becomes so bad. | > | >3) We are still checking if IPSTACK has a problem or not. | >It will take some time to get this result. | >IPSTACK is used in AIBO MIND2 and it is a core object for | >network communication, especially for AIBO Remote Framework. | >And we have tested IPSTACK in oppressive conditions. | > | >Thank you, | > | >AIBO SDE Support | |
| +- 1592 RE:Bug Report for IPSTACK object openrsupport@(Administrator) [2005/01/21-10:11] |
| Response : Edit : Delete |
|
We
apologize for the delay of our answer. 1) This delaying is not a bug. It's used to avoid congestion of small data packets. -> See Nagle's algorithm. There will be a 200 msec delay (maximum) when a program sends a small sized packet. When a program sends a small sized packet, it can be delayed until several other send requests or an ACK packet from the receiver's side is sent. 2) But in OPEN-R, there is a special coding to avoid this delay in the sensor data transmission, which is mainly used for RP OPEN-R. This special code works with data packets that are less than 1024 bytes. As a result, when you send packets over 1024 bytes, you will get the first 1024 bytes soon (using the special code) and get the remaining data after 200 msec. The remaining data is handled by Nagle's algorithm. In the case of the reported test program, there is no send request for 2 sec, so the maximum delay time depends on when the ACK packet from the receiving side is sent. In some operating systems, such as some variations of UNIX, programmers can enable/disable this delay. For example, using the TCP_NODELAY socket option. The maximum timing difference between immediate ACK and delayed ACK also depends on your environment, however a 200 msec timer is used in many implementations. Thank you, AIBO SDE Support (The article that 2005/01/21-10:01 posted, was modified by Administrator) |
| V 1586 tilt and pan center bpaul [2005/01/17-22:59] |
| | Response : Edit : Delete |
|
| I
was looking at Sony's Outside specifications of Aibo ERS-7. I would
like to | know if the tilt and the pan centers of the joint j1 and j2 of the upper | limb(leg) are the same or not. In the document i only see the tilt center and | not the pan one. | My intention is to use one of them as an orgin point for further modeling. | | |
| +- 1587 RE:tilt and pan center ejt [2005/01/18-03:54] |
| | Response : HomePage : Email : Edit : Delete |
|
| >I
was looking at Sony's Outside specifications of Aibo ERS-7. I would
like to | >know if the tilt and the pan centers of the joint j1 and j2 of the upper | >limb(leg) are the same or not. In the document i only see the tilt center and | >not the pan one. | >My intention is to use one of them as an orgin point for further modeling. | I believe they are the "same", if by "same" you mean the axes of rotation | intersect ;) | (For what it's worth, I call them the rotator and elevator joints.) | | If you're modeling the body, you might find this useful, since I just finished | doing some work with that: | http://www.tekkotsu.org/Kinematics.html | |
| +- 1588 RE:tilt and pan center bpaul [2005/01/18-20:03] |
| Response : HomePage : Edit : Delete |
|
>I
believe they are the "same", if by "same" you mean the axes of rotation >intersect ;) Yea thanks what i meant. The tilt axis for the face lies a little above that. >(For what it's worth, I call them the rotator and elevator joints.) Good use of Terminology ;) Also tilt/pan can be confusing for a beginner. >If you're modeling the body, you might find this useful, since I just finished >doing some work with that: >http://www.tekkotsu.org/Kinematics.html Already had a look at it. Their kinmatics model involves the camera and mine does not. Plus they use Denavitt-Hartenberg method of computing kinematic parameters. Same is used by the German team RoboCup people. I think till now no other method has been used. Anyway I can start with specifying the posstion and orientation of the end effector coordinates in 3D and then move on further.. |
| V 1553 get file created time using OPEN-R SDK samlu [2004/12/19-17:24] |
| | Response : Edit : Delete |
|
| I created files using fwrite() API. How can I get the created | time of these files using OPER-R SDK? | I have tried to use stat() to get the file info. It seems | that the time fields are undefined. | |
| +- 1584 RE:get file created time using OPEN-R SDK openrsupport@(Administrator) [2005/01/14-11:40] |
| Response : Edit : Delete |
|
Unfortunately
there are some limitation when using stat(). Please see Section A.1 File system in the Programmer's Guide document. AIBO SDE Support |
| V 1580 How can I run start-rp-openr script? aibo_ido [2005/01/11-21:01] |
| | Response : Edit : Delete |
|
| Hi! | | I have a problem with running start-rp-openr script. I study | SDK_Tutorial7Web_E.pdf on page 26. When I | typing /usr/local/OPEN_R_SDK/RP_OPEN_R/bin/start-rp-openr Cygwin return me the | error: "bash: start-rp-openr: command not found". What command in bash shell I | need to run this script? | |
| +- 1583 RE:How can I run start-rp-openr script? openrsupport@(Administrator) [2005/01/14-09:39] |
| Response : Edit : Delete |
|
Hi, Please make sure to read the Installation Guide to setup Remote Processing OPEN-R. Section 2.3 should provide you with the help that you need. Please let us know if you continue to have problems. Thank you, AIBO SDE Support |
| V 1582 URBI: A new open source alternative to Aibo Programming jcbaillie [2005/01/14-05:18] |
| Response : HomePage : Email : Edit : Delete |
|
Hi, URBI (Universal Robotics Body Interface) is a new way of programming the Aibo robots. It is based on a client/server architecture which makes it possible to program the Aibo using Java or C++ on the client side, or even a simple telnet client. URBI programs can also be run onboard if needed. One example of URBI code to do the classical "Ball Tracking Head": ************************************** whenever (camera.ballx != -1) { headPan.val = headPan.val + camera.xfov * ( 0.5 - camera.ballx ) & headTilt.val = headTilt.val + camera.yfov * ( 0.5 - camera.bally ) }; // ballx and bally are two variables that are constantly set to // the value of the ball position in the image or -1 otherwise *************************************************************** You don't need to known OPENR to use URBI. Useful functions (written in URBI) to make the robot walk, stand, etc will be provided soon in future releases. You will find all the details on the official web page: http://urbi.sourceforge.net The current version (0.9) includes all the features of the official URBI Specification except the function definition. You can access the specification here: http://uei.ensta.fr/baillie/assets/URBI-Specif.pdf If you just want to use the URBI server, simply download the memorystick content here: http://prdownloads.sourceforge.net/urbi/urbiserver-aibo-memorystick-0.9.zip?download You will need the library if you want to use a convenient TCP/IP interface (however, a simple telnet may be enough to start with). The library is available in C++ (for linux or windows) and JAVA. A "liburbi-OPENR" version will be released soon, which will give the possibility to transparently compile standard liburbi-cpp programs to put onboard the robot. Liburbi is available for download here: http://sourceforge.net/projects/urbi/ This is a beta release before the final 1.0 release. Feel free to report bugs and ask questions on the forums: http://sourceforge.net/forum/?group_id=106787 URBI Server and liburbi are released under the GNU GPL. The URBI Language specification is released under a specific license (see inside the archives). Best Regards, Sincerely, Jean-Christophe Baillie (The article that 2005/01/13-21:58 posted, was modified by Author) |
| V 1581 MinimalAddressMgr::DeleteRegion: invalid base ejt [2005/01/12-08:09] |
| Response : HomePage : Email : Edit : Delete |
|
I get a message: MinimalAddressMgr::DeleteRegion: invalid base during boot, but I'm having a hard time tracking down the cause. I'm having a hard time tracking down that error message, or any significant substrings, within any of the source files or even binaries (!?) It doesn't seem to be causing trouble, and I only just noticed it mixed in with other boot messages, so I'm not sure when it first started appearing (and thus what I had last modified) Any idea what's causing it? Thanks! -ethan |
| V 1578 Limitations of the available code slayler [2005/01/10-07:23] |
| | Response : Edit : Delete |
|
| I
wanted to add some functionality to my aibo. From what I see in the | available SDKs, I can't really add anything at all. I would like my aibo to | tell me the local weather forecast in response to speech input. However, this | kind of functionality does not seem possible. The only thing I can do is make | the aibo walk, etc. I'd also like to be able to modify some of the other | Internet functions, such as checking additional email accounts and reading me | the new emails. | | I'm not downplaying the capabilities of the robot out of the box. However, it | seems like Sony is keeping a pretty tight leash (no pun intended) on code that | would really make the aibo useful. | | Anyone have any comments or suggestions? | | I also see a lot of warnings about only using a Sony memory stick reader for | reading and writing to the memory sticks. Is this true. I mean come on. After | paying $2000 for the robot, why would I pay Sony another $60 for their | reader/writer. Has anyone been using another brand of reader/writer without | problems? | |
| +- 1579 RE:Limitations of the available code AiboPet [2005/01/11-02:26] |
| Response : HomePage : Edit : Delete |
|
[I
assume you have an ERS-7 model for the sake of this discussion] >I wanted to add some functionality to my aibo. From what I see in the >available SDKs, I can't really add anything at all. The tools are there. It takes some work. Perhaps the hardest decision is which toolset to pick (RCODE, RFW, OpenR SDK) > I would like my aibo to >tell me the local weather forecast in response to speech input. However, this >kind of functionality does not seem possible. AIBO does not have arbitrary voice recognition. You can capture the audio (using RFW or OpenR SDK) and perform voice recognition on the PC. That's what BrainBo/HomieGate does (ie. AIBO is only part of the system, a lot more processing goes on at the PC) http://www.aibohack.com/210/brainbo.htm http://www.aibohack.com/homie [ERS-7 capable version still being worked on] > The only thing I can do is make >the aibo walk, etc. I'd also like to be able to modify some of the other >Internet functions, such as checking additional email accounts and reading me >the new emails. That can be done rather easily with RFW (AiboMind2) - handling sensors, and talking back. Of course, you must write the mail/internet specific funtions and text scraping routines yourself. Your PC program must initiate "remote control" mode, then you can use the sensors on the robot, or built-in voice commands. Text-to-speech is supported in the RFW SDK, or you can use the standard Windows services. >I'm not downplaying the capabilities of the robot out of the box. However, it >seems like Sony is keeping a pretty tight leash (no pun intended) on code that >would really make the aibo useful. I agree (and have complained about this for some time). However your list of things where they don't release source code doesn't match my list ;-> [ie. I'd prefer RFW and OpenR SDK to work together, but that doesn't address most of the problems you mention] ---- >I also see a lot of warnings about only using a Sony memory stick reader for >reading and writing to the memory sticks.... > ...Has anyone been using another brand of reader/writer without >problems? Sony Brand memory stick readers are now only $20. STRONGLY STRONGLY RECOMMENDED ! http://www.aibohack.com/pms.htm I put all those warnings on my website for good reason, and many past disasters that continue to happen [it spoiled one kid's Xmas present very recently] Using anything else is a risk. There are cases where people have used non-Sony brand readers without problems. There are also many cases where people put a pink stick into a non-Sony brand reader for 1 second (even with the Protect tab turned on), and it wipes out the copy protection ! Since a replacement PMS stick costs $40, and a replacment AiboMind stick costs ~$80 (and other sticks for 210/220/31x models are next-to-impossible to replace), paying $20 for a Sony brand stick reader is a good investment. |
| V 1572 Newbie: Wireless LAN with DHCP bugman_2000 [2005/01/04-00:38] |
| | Response : Email : Edit : Delete |
|
| Hello, | | I am new to Aibo programming and also fairly green in networking. If there is | another thread relating to this question, I'd be grateful if somebody would | point me to it. | | I am currently trying to run the HelloWorld sample OPEN-R program on my Aibo | ERS-7. | | With the AiboMind 2 stick in, my WLAN connection works fine. I copied the | WLANCONF.TXT directly from the AiboMind 2 stick to the blank programming stick | (ERA-MSO16, in directory /OPEN-R/SYSTEM/CONF), so the ESSID, etc. should be | right. DHCP is turned on. | | The manual says to telnet or ping to the Aibo's ISP, but with DHCP on, I can't | figure out how to find it, so I have no idea whether I'm connected. My Aibo | WLAN manager is not finding the Aibo with the empty stick, but I'm not sure | that it should. Can anybody offer any tips on how to proceed? | | I would prefer not to reconfigure the connection to not use the DHCP, since | things are working so well with the Aibo Mind 2 stick. | | Thanks very much, | | Tony | |
| +- 1575 RE:Newbie: Wireless LAN with DHCP AiboPet [2005/01/04-03:01] |
| | Response : HomePage : Edit : Delete |
|
| >>I
am currently trying to run the HelloWorld sample OPEN-R program on my
Aibo | >ERS-7. | | BTW: I suggest starting with a program that does something visible on the dog | (BlinkingLED7). You can never tell if HelloWorld is running (unless the debug | console is working first) | | Depending on which base Open-R software you copied to the memory stick, you | may or may not have a WiFi enabled image. | | You need to copy "WCONSOLE" | If you copied "BASIC" then you have no LAN | If you copied "WLAN" then you have LAN, but the program won't do anything | visible | | ---- | >With the AiboMind 2 stick in, my WLAN connection works fine. I copied the | >WLANCONF.TXT directly from the AiboMind 2 stick to the blank programming stick | >(ERA-MSO16, in directory /OPEN-R/SYSTEM/CONF), so the ESSID, etc. should be | >right. DHCP is turned on. | | With DHCP, you can find the IP address using "Univeral Plug And Play" | Use the Aibo WLAN Manager (V2) and the "AIBO Search" function | | First try it with the AiboMind2 stick. If it works, then try it with the other | stick | | >...I would prefer not to reconfigure the connection to not use the DHCP | FWIW: I find having a hard-coded IP address easier for programming/testing | (ie. you can hard code IP addresses to ping and telnet into batch files and PC | programs) | |
| +- 1576 RE:Newbie: Wireless LAN with DHCP bugman_2000 [2005/01/04-10:10] |
| Response : Email : Edit : Delete |
|
Hi,
Thanks a million for the quick response. >>>I am currently trying to run the HelloWorld sample OPEN-R program on my Aibo >>ERS-7. > >BTW: I suggest starting with a program that does something visible on the dog >(BlinkingLED7). You can never tell if HelloWorld is running (unless the debug >console is working first) Thanks very much. I tried BlinkingLED7 instead, and sure enough, it seems to work. I'm still having trouble finding the Aibo on the network though. >Depending on which base Open-R software you copied to the memory stick, you >may or may not have a WiFi enabled image. > >You need to copy "WCONSOLE" I copied WCONSOLE/memprot/OPEN-R (a few things get overwritten when I copy two different OPEN-R directories to the same stick, but I assume this is okay). I copied them as is, as they were when I downloaded them. Is there anything there I should need to "make" or compile? >With DHCP, you can find the IP address using "Univeral Plug And Play" >Use the Aibo WLAN Manager (V2) and the "AIBO Search" function > >First try it with the AiboMind2 stick. If it works, then try it with the other >stick I am able to find the Aibo using the "AIBO Search" when the AiboMind2 stick is in, but with the programmable stick, using the WLANCONF.TXT file copied directly from the AiboMind2 stick, nothing shows up when I run the AIBO Search. >>...I would prefer not to reconfigure the connection to not use the DHCP >FWIW: I find having a hard-coded IP address easier for programming/testing >(ie. you can hard code IP addresses to ping and telnet into batch files and PC >programs) This makes sense, I may change to that later. Tony |
| V 1567 Compiling Aibo RFW Beta 1 with Borland C++ Builder 6.0 randeep [2004/12/26-04:19] |
| Response : Edit : Delete |
|
Hi, Has anyone successfully compilied (and run) the sample applications that come with Aibo RFW Beta 1. I get linker errors that I am not able to remove even when producing Borland Builder compatible libs using implib utility. The program gives errors like: [Linker error]: Unresolved external VAIBO::Connect(...) from C:\...\Unit1.obj Thanks, Randeep |
| V 1556 reading R.Code ODA files with MoNet pakitomr [2004/12/20-21:58] |
| | Response : Email : Edit : Delete |
|
| I am
trying to read the SYSMTN.ODA file from R-Code distribution (Redist7)
with | the sample aplication MoNet, but it fails. Is this file encripted? It seems to | be. Does it means I can not use the movements created for R-Code in my Open-R | applications? | | Can anybody help me? | | Thanks in advance, Paco. | |
| +- 1558 RE:reading R.Code ODA files with MoNet AiboPet [2004/12/21-02:32] |
| | Response : HomePage : Edit : Delete |
|
| >I
am trying to read the SYSMTN.ODA file from R-Code distribution
(Redist7)... | | Files in \OPEN-R\MW\DATA\E are encrypted (E for Encrypted). SYS*.ODA etc | Files in \OPEN-R\MW\DATA\P are not (P for plain). MOTION.ODA, AUDIO.ODA etc | | Unencrypted files are sometimes GZIP compressed (ie. rename to AUDIO.ODA.gz | and open with WinZip first). You can steal motions from there. | | For other reasons you probably don't want to look inside SYSMTN.ODA (motions | are usually in smaller MTN pieces) | | ---- | >Does it means I can not use the movements created for R-Code in my Open-R | >applications? | | No, it means you can't use the built-in system motions (easily) with Open-R. | For most things this is not a limitation. | | If you find a really good builtin motion you like, please let me know (I'm | particularly fond of the AiboMind2 ball playing on back or pole skits...) | | --- | New movements created for RCODE (created with MEdit or Skitter) will work fine. | |
| +- 1562 RE:reading R.Code ODA files with MoNet pakitomr [2004/12/21-18:40] |
| | Response : Email : Edit : Delete |
|
| >No,
it means you can't use the built-in system motions (easily) with Open-R. | >For most things this is not a limitation. | | >If you find a really good builtin motion you like, please let me know (I'm | >particularly fond of the AiboMind2 ball playing on back or pole skits...) | | I am interested in walk and turn motions, with different angles and distances | (basically the ACTION:WALK:* y ACTION:TURN:*). I think this motions are good | enough for the work I do. | | Do you know where I can obtain these motions? Do you know if anybody has created | free motions for walk and turn availables in any place? | |
| +- 1563 RE:reading R.Code ODA files with MoNet AiboPet [2004/12/22-02:19] |
| | Response : HomePage : Edit : Delete |
|
| >I
am interested in walk and turn motions, with different angles and
distances | >(basically the ACTION:WALK:* y ACTION:TURN:*). I think this motions are good | >enough for the work I do. | | The WALK/TURN actions are not stored as motion files. | They are done with an algorithmic walk engine (LG.BIN) | | ---- | There is a library called OMWares that provides walking for the ERS-210/220 | *only*, part of the SoccerLion example (using an older version of LG.BIN, that | walks similar to AiboLife1) | | The Sony folks have decided not to port that to the ERS-7 which is unfortunate | (see past posts and ignored pleas for an ERS-7 version of OMWares) | | Currently the only ERS-7 capable walk engines are the ones from RoboCup -- | which usually have AIBO walking on his knees with greater wear-and-tear (The | CMUPack02 winning source code is one such example - that's what I use in | WalkBench & AiboSpeedRacer, and also used in Tekkotsu IIRC) | | http://www.aibohack.com/openr_sdk | | If you want AIBO to walk normally on the ERS-7, then I suggest using RCODE or | RFW (they do have a working LG module, but not OpenR SDK compatible) | |
| +- 1565 RE:reading R.Code ODA files with MoNet ejt [2004/12/23-00:52] |
| Response : HomePage : Email : Edit : Delete |
|
>CMUPack02
winning source code is one such example - that's what I use in >WalkBench & AiboSpeedRacer, and also used in Tekkotsu IIRC) We have a walk editing tool in Tekkotsu which will let you create an upright walk... walks are surprisingly sensitive to small changes in parameters, so it's nice to have realtime feedback like our editor provides. In my intuitive feel from tweaking walks, it's hard to get a smooth walk that is "off the knees" because of the way the weight is distributed - the head pulls the center of gravity fairly far forward, which means the front legs have to reach out to center it, which means it winds up walking on the forearms instead of the paws. In any case, a knee walk isn't necessarily bad, it's just the speed you drive it at that will wear the motors (not that we've had any problem with that yet though) You can just increase the period of the walking cycle and not tell it to walk as quickly to reduce strain, regardless of the posture of the walk. Anyway, the CMPack'02 engine is quite flexible and capable of supporting other gaits, for instance a 'pace' gait shown here: (second movie) http://www-2.cs.cmu.edu/%7Etekkotsu/Samples.html#walking However, no one in our group has had time to tweak up a non-knee walk - if anyone is interested in working on one, we'd love to include it in our releases. A 'crawl' gait (only one leg in the air at a time) would also be nice, either knee or non-knee, for times when precision is better than speed. Anyone? Anyone? Bueller? ;) |
| V 1555 Wireless LAN Card for an ERS-210 bradleyjs [2004/12/20-04:33] |
| | Response : Edit : Delete |
|
| Does
anyone know where I can purchase a LAN card for my ERS-210? Can't seem
to | find one available; assuming no longer in production. | |
| +- 1559 RE:Wireless LAN Card for an ERS-210 AiboPet [2004/12/21-02:37] |
| | Response : HomePage : Edit : Delete |
|
| >Does
anyone know where I can purchase a LAN card for my ERS-210? Can't seem
to | >find one available; assuming no longer in production. | | They are officially discontinued everywhere outside of Japan. | | Sometimes you can find them on eBay. Either the official "ERA-201D1" or the | VAIO work-alike "PCWA-C150S" (other models will not work) | | Cross your fingers. They are rare. | | ---- | The only (non-second-hand) place selling them is the Japanese AIBO website | http://www.jp.aibo.com/products/ERA201D1.html | They only sell to Japan, so you need to find an importer. | Around $100 for the card, but import fees and shipping will close to double | the price. | |
| +- 1561 RE:Wireless LAN Card for an ERS-210 bradleyjs [2004/12/21-05:56] |
| Response : Edit : Delete |
|
Thanks.
The first place I searched was eBay and it didn't seem to find any hits for the ERA-201D1. I'll try the other card and see what happens. >>Does anyone know where I can purchase a LAN card for my ERS-210? Can't seem to >>find one available; assuming no longer in production. > >They are officially discontinued everywhere outside of Japan. > >Sometimes you can find them on eBay. Either the official "ERA-201D1" or the >VAIO work-alike "PCWA-C150S" (other models will not work) > >Cross your fingers. They are rare. > >---- >The only (non-second-hand) place selling them is the Japanese AIBO website >http://www.jp.aibo.com/products/ERA201D1.html >They only sell to Japan, so you need to find an importer. >Around $100 for the card, but import fees and shipping will close to double >the price. |
| V 1554 Bug Report in R-Code pakitomr [2004/12/19-18:13] |
| | Response : Email : Edit : Delete |
|
| English
version of R-code has a bug when reading from the legs values. When you | want read the RR Leg, the RF values are returned. The japanese version is ok $B%=(B?. | |
| +- 1560 RE:Bug Report in R-Code AiboPet [2004/12/21-02:39] |
| Response : HomePage : Edit : Delete |
|
>English
version of R-code has a bug when reading from the legs values. When you >want read the RR Leg, the RF values are returned. The japanese version is ok $B%=(B?. Easy to fix yourself: edit the \OPEN-R\APP\PC\AMS\JOINT.CFG file The "RRLeg_1/2/3" numbers are incorrectly numbered 41, 42, 43 (should be 51, 52, 53) |
| V 1543 how to control the joints correctly totoduke [2004/12/02-05:57] |
| | Response : Edit : Delete |
|
| I am
working with ERS-7 for a while. But there is also a problem when I want | to control the joints of the head and legs. When I control the head to move | to some angles, the legs of the dogs are shivering strongly. But I didn't | input any joint value to the legs. And when I want to control the leg joints | (no values to the head joints), the head is also shivering. Did I miss | something when controlling both legs and head? | |
| +- 1557 RE:how to control the joints correctly pakitomr [2004/12/20-22:03] |
| Response : Email : Edit : Delete |
|
Try
to fill the values for all the joints |
| V 1546 GetRobotDesign() for ERS-7M2 ejt [2004/12/14-06:24] |
| | Response : HomePage : Email : Edit : Delete |
|
| Just
wondering whether the M2 models will return "ERS-7M2" for their robot | design string, or will just stick with "ERS-7" since there aren't any hardware | differences (or are there??) | | -ethan | |
| +- 1552 RE:GetRobotDesign() for ERS-7M2 AiboPet [2004/12/16-01:40] |
| Response : HomePage : Edit : Delete |
|
>
... or will just stick with "ERS-7" since there aren't any hardware >differences (or are there??) Correct it stays "ERS-7". BTW: the "ERS-7" string is in the software (DESIGN.CFG) keyed off the CPC codes. Hardware differences for the AIBO - black paint and grey ears (and I assume newer / fixed leg servos). The M2 charging station has a different mat (b&w versus color) |
| V 1541 Error with compiling ImageCapture host aibo_ido [2004/12/01-19:35] |
| | Response : Edit : Delete |
|
| Hi! | I'm going to know OPEN-R programming to AIBO 220 model. I'm studying book | SDK_Tutorial1-7Web.pdf and I have a problem with "make install" comand in | SDK_Tutorial7Web.pdf, page 22, when I compile ImageCapture/host/makefile. The | error is "...cannot specify -o with -c or -S and multiple compilations..." | typed by Cygwin. What should I do? | |
| +- 1547 RE:Error with compiling ImageCapture host openrsupport@(Administrator) [2004/12/14-12:58] |
| | Response : Edit : Delete |
|
| Which
'make install' on page 22 causes problems? There are thee 'make install' | commands on that page. | | Can you please tell us the command line before (previous line)"...cannot specify | -o with -c or -S and multiple compilations..."? That line should be causing the | error. | | Thank you, | | AIBO SDE Support | |
| +- 1549 RE:Error with compiling ImageCapture host aibo_ido [2004/12/14-20:59] |
| | Response : Edit : Delete |
|
| >Which
'make install' on page 22 causes problems? There are thee 'make install' | >commands on that page. | > | >Can you please tell us the command line before (previous line)"...cannot | specify | >-o with -c or -S and multiple compilations..."? That line should be causing | the | >error. | > | >Thank you, | > | >AIBO SDE Support | | The full error is "g++ -ggdb3 -I. -I.. '/usr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp- | openr-config --cflags' -o ImageObserverStub.o -c ImageObserverStub.cc g++: | cannot specify -o with -c or -S and multiple compilations. make: *** | [ImageObserverStub.o] error 1". I wrote the makefile as it's need in | path /host. Also I've done the PMS by command "make install" in path /robot. | What should I do? | | Thank you. | |
| +- 1550 RE:Error with compiling ImageCapture host openrsupport@(Administrator) [2004/12/15-10:20] |
| Response : Edit : Delete |
|
There
appears to be a simple typo in you makefile. ' is not the same as ` BAD: CXXFLAGS= \ -O2 \ -g \ -I. \ -I.. \ '$(OPENRCONFIG) --cflags' GOOD: CXXFLAGS= \ -O2 \ -g \ -I. \ -I.. \ `$(OPENRCONFIG) --cflags` Thank you, AIBO SDE Support |
| V 1545 Cannot compile/run RFW in Visual C++ 6.0 on Windows XP SP1 randeep [2004/12/13-04:12] |
| | Response : Edit : Delete |
|
| Hi, | | While compiling RFW in VC 6, i get an error of cannot link file ij115.lib but it | is able to take other *.lib files !!! in the release version. In the debug | version I am able to get a file *.exe (like SimpleNavi.exe) but when i execute | it, it says: | | The procedure entry point _chkesp could not be located in the dynamic link | library MSVCRTD.dll. | | Please help me out. | | Thanks, | Randeep | |
| +- 1548 RE:Cannot compile/run RFW in Visual C++ 6.0 on Windows XP SP openrsupport@(Administrator) [2004/12/14-13:05] |
| Response : Edit : Delete |
|
Please
check the Link options in the VC++ project. The ijl15.lib file is specified as ../../lib/release/ijl15.lib in the link library setting. See (2) in https://openr.aibo.com/openr/eng/perm/rfw/regi_faq3.php4#B0003 ID:B0003 Q.How do I setup the development environment for AIBO Remote Framework? If you have experience with VC++, you can change the lib folder. >The procedure entry point _chkesp could not be located in the dynamic >link library MSVCRTD.dll. I'm not sure what this problem is right now. I did an Internet search for "_chkesp and MSVCRTD.dll". _chkesp is a call stack check function. MSVCRTD.dll (debug version of MSVCRT.dll) doesn't have that symbol. You may be able to get an answer from a VC++ user's website. AIBO SDE Support |
| V 1539 Remote Framework support for ERS-220 ejfuller [2004/11/30-01:13] |
| | Response : Email : Edit : Delete |
|
| Does
anyone know if Sony is planning to support the ERS-220 with the Remote | Framework? The current version of Remote Framework only supports ERS-7, but | older versions of the Beta for Remote Framework used to support ERS-220 models. | |
| +- 1540 RE:Remote Framework support for ERS-220 AiboPet [2004/11/30-02:07] |
| Response : HomePage : Edit : Delete |
|
>..
older versions of the Beta for Remote Framework used to support ERS-220 models. Minor correction: The RFW Beta was ERS-7 only as well ;-< I too would like to see an RFW for the older 210/220 models. Perhaps without the proprietary features like Visual Pattern Recognition (licensed from a separate company). |
| V 1534 How to detect the distance exactly? wisdom [2004/11/26-14:36] |
| | Response : Email : Edit : Delete |
|
| Hello ! | The positioning module to mobile robot should be based on a relatively | accurate distance - | detecting platform.I just want to implement the array drill of Aibo.Any | suggestions?Thanks a lot. | (The article that 2004/11/26-12:18 posted, was modified by Author) | |
| +- 1538 RE:How to detect the distance exactly? AiboPet [2004/11/29-03:48] |
| Response : HomePage : Edit : Delete |
|
> Any suggestions? Unclear what you are asking for, and for what model robot, and what tools. RCODE: Rcode has a 'Distance' variable (2x0 and ERS-7) and 'Distance_Cliff' (ERS-7 only) that give the distance to obstacle in millimetres Open R SDK: I suggest looking at the 'SensorObserver' OpenR SDK sample. In particular the "PSD" sensors (1 on the ERS-2x0, 3 on the ERS-7) At a hardware level, they use Sharp IR distance sensors AFAIK |
| V 1478 Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION ASAP! wangy22 [2004/11/07-14:28] |
| | Response : Edit : Delete |
|
| I am
using Linux and want to write software for ERS-7 using OPEN-R SDK | | After I copied /usr/local/OPEN_R_SDK/OPEN_R/MS_ERS7/WLAN/memprot/OPEN-R to | /mnt/sda1, which is the path to the memory stick, I modified the wlanconf.txt | file to the following information: | ########################### | HOSTNAME=AIBO | ETHER_IP=10.0.1.2 | ETHER_NETMASK=255.255.255.0 | IP_GATEWAY=10.0.1.2 | ESSID=AIBONET | WEPENABLE=0 | WEPKEY=AIBO2 | APMODE=0 | CHANNEL=3 | #DNS_SERVER_1=10.0.1.1 | #DNS_DEFDNAME=example.net | USE_DHCP=0 | ########################### | | The problem is when I ping 10.0.1.2, I just get: | ########################### | $ ping 10.0.1.2 | PING 10.0.1.2 (10.0.1.2) 56(84) bytes of data. | From 10.0.1.1 icmp_seq=2 Destination Host Unreachable | From 10.0.1.1 icmp_seq=3 Destination Host Unreachable | From 10.0.1.1 icmp_seq=4 Destination Host Unreachable | ########################### | | Here, 10.0.1.1 is the IP of my computer. I also have another computer in the | wireless network, whose IP is 10.0.1.3 | | I have already set their essid to AIBONET and set the mode to ad-hoc | | Now I can ping my another computer(10.0.1.3) in the wireless network, but just | cannot ping 10.0.1.2 | | Any ideas? | |
| +- 1480 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION ASAP! AiboPet [2004/11/08-00:59] |
| | Response : Edit : Delete |
|
| "Unreachable"
is a different problem than "not found", and usually means an IP | addressing error, not AIBO specific | I suspect 10.0.1.2 is used by something else (eg: gateway ??) | | >Any ideas? | Use 10.0.1.100 instead. | | I also recommend people use larger numbers for their PC (eg: 10.0.1.90 for PC, | 10.0.1.100 for AIBO) to avoid these kinds of problems. | |
| +- 1482 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS wangy22 [2004/11/08-09:30] |
| | Response : Edit : Delete |
|
| I've
tried your suggestion, but it doesn't work either. | | I think it is the problem of the configuration in AIBO, not other computers in | the wireless network, since they can ping each other correctly in ad-hoc mode. | | Now I've got a wireless router and I can ping AIBO and even telnet to AIBO after | I set APMODE to 1 in the wlanconf.txt file. So that means the wireless adapter | in AIBO is working fine. | | But that is not what I want. Since not everywhere has a wireless router, I | really want to get ad-hoc mode wireless working. | | Is there anyone using ad-hoc mode? Can you post your wlanconf.txt? | | Thanks! | | >"Unreachable" is a different problem than "not found", and usually means an IP | >addressing error, not AIBO specific | >I suspect 10.0.1.2 is used by something else (eg: gateway ??) | > | >>Any ideas? | >Use 10.0.1.100 instead. | > | >I also recommend people use larger numbers for their PC (eg: 10.0.1.90 for PC, | >10.0.1.100 for AIBO) to avoid these kinds of problems. | |
| |- 1485 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS AiboPet [2004/11/09-01:34] |
| | | Response : Edit : Delete |
|
| | >>
APMODE=0 | | Sorry I missed this the first time. | | | | Don't set the mode to 0, keep it to 2 (automatic) for all uses | | APMODE=2 | | | | It will automatically work in either Infrastructure (when within range of an | | access point) or Ad-Hoc. | | | | In the Ad-Hoc case, the PC WiFi card | | + must be setup to Ad-Hoc as well | | + must be on the same channel | | + same SSID, WEP etc of course | | | | AND MOST IMPORTANT: | | + must be running BEFORE you boot AIBO (ie. Link light on or flashing) | | |
| | +- 1505 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS wangy22 [2004/11/15-11:26] |
| | | Response : Edit : Delete |
|
| | >>>
APMODE=0 | | >Sorry I missed this the first time. | | > | | >Don't set the mode to 0, keep it to 2 (automatic) for all uses | | >APMODE=2 | | > | | >It will automatically work in either Infrastructure (when within range of an | | >access point) or Ad-Hoc. | | > | | >In the Ad-Hoc case, the PC WiFi card | | >+ must be setup to Ad-Hoc as well | | >+ must be on the same channel | | >+ same SSID, WEP etc of course | | > | | >AND MOST IMPORTANT: | | >+ must be running BEFORE you boot AIBO (ie. Link light on or flashing) | | | | Thanks, that solved my problem. | | |
| | +- 1536 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS ioneer [2004/11/27-19:27] |
| | Response : Edit : Delete |
|
| so
what should i do if i have XP and aibo and have the same problem? they | don't want to connect even when i follow all the points in this thread | how you solved it? | | >>>> APMODE=0 | >>Sorry I missed this the first time. | >> | >>Don't set the mode to 0, keep it to 2 (automatic) for all uses | >>APMODE=2 | >> | >>It will automatically work in either Infrastructure (when within range of an | >>access point) or Ad-Hoc. | >> | >>In the Ad-Hoc case, the PC WiFi card | >>+ must be setup to Ad-Hoc as well | >>+ must be on the same channel | >>+ same SSID, WEP etc of course | >> | >>AND MOST IMPORTANT: | >>+ must be running BEFORE you boot AIBO (ie. Link light on or flashing) | > | >Thanks, that solved my problem. | |
| +- 1486 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS curie [2004/11/09-02:35] |
| | Response : Edit : Delete |
|
| The
terminology is confusing. | "Ad-hoc mode" that AIBO supports with APMODE=0 is OLD ad-hoc mode (ad-hoc demo | mode). | "Ad-hoc mode" of Windows XP is different thing (IBSS peer-to-peer mode) | that is compatible with infrastructure mode. | |
| +- 1535 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS ioneer [2004/11/27-19:26] |
| | Response : Edit : Delete |
|
| so
what should i do if i have XP and aibo and have the same problem? they | don't want to connect even when i follow all the points in this thread | | >The terminology is confusing. | >"Ad-hoc mode" that AIBO supports with APMODE=0 is OLD ad-hoc mode (ad-hoc demo | >mode). | >"Ad-hoc mode" of Windows XP is different thing (IBSS peer-to-peer mode) | >that is compatible with infrastructure mode. | |
| +- 1537 RE:Can't get ERS-7 ad-hoc wireless working. NEED SOLUTION AS AiboPet [2004/11/29-03:41] |
| Response : HomePage : Edit : Delete |
|
>so
what should i do if i have XP and aibo and have the same problem? they >don't want to connect even when i follow all the points in this thread For general WiFi LAN help, I suggest checking the aibo-life.org forums, especially the FAQs. http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi Exact details will depend on your PC WiFi setup, which model AIBO (ERS-7 has more features than earlier models), which features you want working (basic connection is much easier than AiboMail) etc Once you get a configuration working, save the working WLANCONF.TXT file. --- My abbreviated advice: use the default ERS-210 settings for WLANCONF.TXT (see below), and configure your PC WiFi card around them [no DHCP, no advanced features, etc] --- HOSTNAME=AIBO ETHER_IP=10.0.1.100 ETHER_NETMASK=255.255.255.0 IP_GATEWAY=10.0.1.1 ESSID=AIBONET WEPENABLE=1 WEPKEY=AIBO2 APMODE=2 CHANNEL=3 |
| V 1533 compiling Java for Aibo kolargol00 [2004/11/25-00:20] |
| Response : Edit : Delete |
|
Hello all! Is there any existing way to compile Java programs for the Aibo? I'd really like to use Java to do some serious OOP. Since the OPEN-R SDK uses GNU tools I can imagine using the "GNU Compiler for the Java Programming Language" (GCJ, see http://gcc.gnu.org/java/) to cross-compile Java into native Aibo machine code. Has anyone tried this or knows how to turn GCJ into a cross-compiler for Aibo? Regards, Etienne |
| V 1507 basic behavior bpaul [2004/11/15-23:35] |
| | Response : HomePage : Edit : Delete |
|
| Hello again, | If i implement a full motion module using open-r then would i have to include an | advanced obstacle-detection or can i somehow use the existing ones. | Are there some basic behvior(s) that Aibo implements without any memory stick | inside apart from posture control or sensor data retrieval. | thx | |
| +- 1508 RE:basic behavior AiboPet [2004/11/16-01:17] |
| | Response : Edit : Delete |
|
| >Hello again, | >If i implement a full motion module using open-r then would i have to include | an | >advanced obstacle-detection or can i somehow use the existing ones. | | Sorry no. The OpenR SDK is designed for a ground-up implementation of almost | everything (walking, obstacle detection, etc - all do it yourself) | | There is (OpenR SDK compatible) source code available from a number of RoboCup | teams that provide a good place to start -- walking designed for speed, | colored ball and field marker detection. | | If you would rather have your AIBO act more like a regular dog (normal | walking, reaction to faces and visual patterns etc), I suggest looking at the | RFW (ERS-7 only). It provides most of the features available in the high-end | commercial software (AiboMind) to a programmer - with the program running on | the PC. | | Also available is RCODE/RCodePlus, which has many of these things built in, | but with limited extensibility. Probably not a good fit for your application, | but something to always consider. | | >Are there some basic behvior(s) that Aibo implements without any memory stick | >inside apart from posture control or sensor data retrieval. | Without a memory stick, AIBO will run a very simplified personality for | diagnostics (clinic or demo mode). It is not compatible with any programming | tool (ie. you *need* a memory stick to do anything useful) | |
| +- 1509 RE:basic behavior bpaul [2004/11/16-06:15] |
| | Response : Edit : Delete |
|
| >Sorry
no. The OpenR SDK is designed for a ground-up implementation | That's really the most -ve thing about Open-R | | There is (OpenR SDK compatible) source code available from a number of RoboCup teams that provide | a good place to start -- walking designed for speed, colored ball and field marker detection. | Think i'll try that option | | If you would rather have your AIBO act more like a regular dog (normal | >walking, reaction to faces and visual patterns etc), I suggest looking at the | >RFW (ERS-7 only). It provides most of the features available in the high-end | >commercial software (AiboMind) to a programmer - with the program running on | >the PC. | I wonder where RFW can be found ? | RCODE/RCodePlus,are not options i am considering. | |
| +- 1511 RE:basic behavior AiboPet [2004/11/17-03:08] |
| | Response : Edit : Delete |
|
| >I
wonder where RFW can be found ? | | RFW = abbreviation for "AIBO Remote Framework" - see button at the top of this | page. | | Final version just released. Requires the AiboMind V2 software (running on the | 32MB AiboMind stick) | |
| +- 1515 RE:basic behavior bpaul [2004/11/17-20:09] |
| | Response : HomePage : Edit : Delete |
|
| >RFW
= abbreviation for "AIBO Remote Framework" - see button at the top of
this | >page. | Sorry about not guessing that one | >Final version just released. Requires the AiboMind V2 software (running on the | >32MB AiboMind stick) | I have kept my distance from "rfw" cause of the extra burden of learning windows programming. | | I'm more of a mac freak, and no options of remote-fw exisits there. There is a software called Jam. Built | using R-Util, Jam can connect to and control any Aibo running AiboPet's RCode Plus 2.5x. | Thats all there is to the mac-side, beside open-r prog. | | U ever heard about URBI http://sourceforge.net/projects/urbi/ | |
| +- 1516 RE:basic behavior ejt [2004/11/18-04:00] |
| | Response : HomePage : Email : Edit : Delete |
|
| >I'm
more of a mac freak, and no options of remote-fw exisits there. There
is a | >software called Jam. Built using R-Util, Jam can connect to and control any | >Aibo running AiboPet's RCode Plus 2.5x. | >Thats all there is to the mac-side, beside open-r prog. | If you're going to be "stuck" with open-r, you might want to take a look at our | Tekkotsu framework: | http://www.tekkotsu.org/ | It doesn't have as much of the high level stuff (yet ;) as RFW, but it'll give | you a fair amount of included behaviors and development utilities. | | I'm a mac nut too btw ;) We use Linux in the lab, but I also do a fair bit of | development from home or on the road on my iBook, so you can be assured | everything will/should run smoothly - for example all of the GUI utilities are | in Java, command line tools are platform independent, etc. | (The article that 2004/11/18-03:57 posted, was modified by Author) | |
| +- 1532 RE:basic behavior AiboPet [2004/11/24-01:24] |
| Response : HomePage : Edit : Delete |
|
>>I'm
more of a mac freak, and no options of remote-fw exisits there.... Apparently the RFW DLLs *do* work on the Mac with Virtual PC software (ie. the Aibo Entertainment Player works). The free C++ command line compiler should work as well. The main reason for using RFW is to get those high level features (for more-or- less free). Voice recognition, visual pattern recognition, large library of walk and motion routines (not the CMU knee walk) - and the added advantage you can merge your software into the running AiboMind2 software [no need for users to swap sticks] The insane amount of time it would take to reinvent those features yourself would easily be saved by going out and buying a Windows box ;-> --- FWIW: I'm working on a trimmed down RFW glue helper for PDAs that would also work for other platforms (ie. removing reliance on VAIBOServer.EXE and the DLLs) |
| V 1521 Anybody for URBI bpaul [2004/11/18-19:19] |
| | Response : HomePage : Edit : Delete |
|
| URBI:
Universal Robotic Body Interface. URBI is a scripted command language
used | to control robots (AIBO, pioneer,...). It is a robot-independant API. Based on a | client/server architecture it is designed to be used over a TCP/IP or IPC | connection. | | Did anyone yet put a hand on that ? | |
| +- 1524 why not ? bpaul [2004/11/20-00:19] |
| | Response : Edit : Delete |
|
| It
seems to function at the same level as OPEN-R. IT promises a lot. Why
isn't | then anybody curious about it ? | PLEASE RESPOND | |
| +- 1526 RE:why not ? AiboPet [2004/11/20-01:16] |
| | Response : HomePage : Edit : Delete |
|
| >
Why isn't then anybody curious about it ? | | I've looked at it (including your latest update) | Speaking only for myself -> | | There are already several tools for programming AIBO at the OpenR SDK level. | The official API, the many class libraries built by the RoboCup competitors, | and Tekkotsu | | Functionality-wise there are very similar - but vary greatly in the 'style' or | programs you write. | Although not done, I see URBI as being yet another way of doing the same old | stuff | | --- | IMHO the two big problems needing solving for AIBO programming: | #1) the gaps in functionality (OpenR SDK too low, RFW and RCODE too high - and | nothing in the middle) | #2) not enough programmers (heck it is tough enough to get people to try | YART...) | |
| +- 1528 RE:why not ? bpaul [2004/11/21-02:09] |
| | Response : Edit : Delete |
|
| >There
are already several tools for programming AIBO at the OpenR SDK level. | >The official API, the many class libraries built by the RoboCup competitors, | >and Tekkotsu | | But none as simple to use as URBI. Personally i think its a great advantage. | > | >Functionality-wise there are very similar - but vary greatly in the 'style' or | >programs you write. Although not done, I see URBI as being yet another way of | doing the same old stuff | | I think to a certain you can achieve in Urbi what you can do using open-r or | tekkotsu, only in less time. Not to mention in future releases it will support | more languages even java. | >--- | >IMHO the two big problems needing solving for AIBO programming: | >#1) the gaps in functionality (OpenR SDK too low, RFW and RCODE too high - and | >nothing in the middle) | Well guess in Robotics, the middel of the road is difficult to find. | >#2) not enough programmers (heck it is tough enough to get people to try | >YART...) | Ask me, i have searched for them, for months now, they are hardly to be found ! | |
| +- 1531 RE:why not ? AiboPet [2004/11/24-01:17] |
| Response : HomePage : Edit : Delete |
|
>But
none as simple to use as URBI. Personally i think its a great advantage. Stating the obvious -> the person inventing a new API always thinks their API is better than the alternatives ;-> Ask ejt about Tekkotsu... --- So to put it another way, you should find someone (not you) who isn't writing AIBO programs right now -- and find out why not. Then see if there is something you can do to make it easier for them. For example many people will not even consider using the CMU 'knee walk' walk engine with their AIBOs, or install cygwin or deal with command line tools, ... |
| V 1498 Aibo Remote Framework on .NET frish [2004/11/12-01:22] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I am just wondering if there is any AIBO Remote Framework for Visual | Studio .NET. I am a C# developer interested in developing some windows | applications for the AIBO. | I know the question has been here before in different other forms but I do not | have a convincing answer yet. | | Thank you, | |
| +- 1501 RE: Aibo Remote Framework on .NET openrsupport@(Administrator) [2004/11/12-10:08] |
| | Response : Edit : Delete |
|
| Thank
you for your interest in AIBO Remote Framework. Currently Visual C++ 6.0 | and Visual C++.NET 2003 are the only environments that are supported by AIBO | Remote Framework. However, it would be possible to create a .NET wrapper DLL | for the C++ DLL, but this would not be officially support by the AIBO SDE. | | https://openr.aibo.com/openr/eng/perm/rfw/regi_faq2.php4#A0004 | | AIBO SDE Support | |
| +- 1529 RE: Aibo Remote Framework on .NET bhchan [2004/11/23-03:38] |
| Response : Email : Edit : Delete |
|
>However,
it would be possible to create a .NET wrapper DLL >for the C++ DLL, but this would not be officially support by the AIBO SDE. Anyone sharing their C# wrapper? |
| V 1518 R-CODE 1.2 for ERS-210 bradleyjs [2004/11/18-08:01] |
| | Response : Edit : Delete |
|
| Ok, what am I doing wrong. I unzipped the file, copied the | RCODE\RCODE210_E\OPEN-R folder to the memory stick, then created a simple | script file called R-CODE.R and nothing happens. | | Here's the contents of the script: | EDIT | FOR:i:1:10 | PLAY:ACTION:STAND | WAIT | NEXT | EXIT | END | | Anybody have any suggestions? | | Am I supposed to apply the "Flash" on the ERS-210 first? | |
| +- 1519 RE:R-CODE 1.2 for ERS-210 bradleyjs [2004/11/18-08:38] |
| | Response : Edit : Delete |
|
| >Ok,
what am I doing wrong. I unzipped the file, copied the | >RCODE\RCODE210_E\OPEN-R folder to the memory stick, then created a simple | >script file called R-CODE.R and nothing happens. | > | >Here's the contents of the script: | > EDIT | > FOR:i:1:10 | > PLAY:ACTION:STAND | > WAIT | > NEXT | > EXIT | > END | > | >Anybody have any suggestions? | | Anybody have any suggestions or a simple test script to start with? | | > | >Am I supposed to apply the "Flash" patch on the ERS-210 first? | |
| +- 1520 RE:R-CODE 1.2 for ERS-210 bradleyjs [2004/11/18-09:42] |
| | Response : Edit : Delete |
|
| Ok,
I found YART and that's working for me. Is there really any big
difference | between the "new" v1.2 release that came out 7 days ago and the one that comes | with YART? | | Thanks. | | >>Ok, what am I doing wrong. I unzipped the file, copied the | >>RCODE\RCODE210_E\OPEN-R folder to the memory stick, then created a simple | >>script file called R-CODE.R and nothing happens. | >> | >>Here's the contents of the script: | >> EDIT | >> FOR:i:1:10 | >> PLAY:ACTION:STAND | >> WAIT | >> NEXT | >> EXIT | >> END | >> | >>Anybody have any suggestions? | > | >Anybody have any suggestions or a simple test script to start with? | > | >> | >>Am I supposed to apply the "Flash" patch on the ERS-210 first? | |
| +- 1522 RE:R-CODE 1.2 for ERS-210 AiboPet [2004/11/19-02:15] |
| | Response : HomePage : Edit : Delete |
|
| >Ok,
I found YART and that's working for me. | You beat me to it => always start with YART ! | | > Is there really any big difference | >between the "new" v1.2 release that came out 7 days ago and the one that comes | >with YART? | | The release that came out 7 days ago is a repackaging of the RCODE 11x and 2x0 | and a little documentation that was on the now discontinued ERA website (ie. | stuff that is 2+ years old) | | --- | The YART release includes RCodePlus (built on RCODE1.2) a few extras like the | YART and YART_APG utilities. | | RCODE 1.2 programs will run on either. | | >>>Ok, what am I doing wrong.... | Don't use EDIT/END when editing the R-CODE.R file directly (see the R-CODE.R | generated/edited by YART for ideas) | | If you connect to the AIBO using telnet, you use the "EDIT" ... "END" process | for uploading a new script. | | >Am I supposed to apply the "Flash" patch on the ERS-210 first? | Not required for RCODE/RCodePlus/YART | If you have a very old ERS-210, you need the flash upgrade to run OPEN-R SDK | programs. | |
| +- 1523 RE:R-CODE 1.2 for ERS-210 bradleyjs [2004/11/19-05:08] |
| | Response : Email : Edit : Delete |
|
| Thanks
a bunch AiboPet. I've been reading the .R file since last night. | | Do you have any additional programming documentation that shows the entire API? | | >>Ok, I found YART and that's working for me. | >You beat me to it => always start with YART ! | > | >> Is there really any big difference | >>between the "new" v1.2 release that came out 7 days ago and the one that comes | >>with YART? | > | >The release that came out 7 days ago is a repackaging of the RCODE 11x and 2x0 | >and a little documentation that was on the now discontinued ERA website (ie. | >stuff that is 2+ years old) | > | >--- | >The YART release includes RCodePlus (built on RCODE1.2) a few extras like the | >YART and YART_APG utilities. | > | >RCODE 1.2 programs will run on either. | > | >>>>Ok, what am I doing wrong.... | >Don't use EDIT/END when editing the R-CODE.R file directly (see the R-CODE.R | >generated/edited by YART for ideas) | > | >If you connect to the AIBO using telnet, you use the "EDIT" ... "END" process | >for uploading a new script. | > | >>Am I supposed to apply the "Flash" patch on the ERS-210 first? | >Not required for RCODE/RCodePlus/YART | >If you have a very old ERS-210, you need the flash upgrade to run OPEN-R SDK | >programs. | |
| +- 1525 RE:R-CODE 1.2 for ERS-210 AiboPet [2004/11/20-01:07] |
| | Response : HomePage : Edit : Delete |
|
| >Do
you have any additional programming documentation that shows the entire | API? | | Look at the 10MB+ RCODE 2x0/11x archive you downloaded - at the | file "rcode/rcode12ref.txt" | | That describes the RCODE 1.2 keywords, variables etc. | The RCodePlus extensions are documented on my website. | http://www.aibohack.com/rcode/rcp_overview.htm | | Mostly reference for keywords. Not a lot of advice for writing the scripts. | After outgrowing YART, for interactively playing around with RCODE, I suggest | getting a WiFi setup and using 'telnet' to type in RCODE/RCodePlus commands | directly. | |
| +- 1527 RE:R-CODE 1.2 for ERS-210 bradleyjs [2004/11/20-09:30] |
| Response : Edit : Delete |
|
Yes,
I've read the rcode12ref.txt a few times now. Just wanted to see your
docs for the extended API. My final purchase will be the wireless LAN card for the 210; already have an wireless LAN in the house (can't live without it!). I can't wait to eventually start writing some C++ routines to extend his capabilities. Thanks for the info! >>Do you have any additional programming documentation that shows the entire >API? > >Look at the 10MB+ RCODE 2x0/11x archive you downloaded - at the >file "rcode/rcode12ref.txt" > >That describes the RCODE 1.2 keywords, variables etc. >The RCodePlus extensions are documented on my website. >http://www.aibohack.com/rcode/rcp_overview.htm > >Mostly reference for keywords. Not a lot of advice for writing the scripts. >After outgrowing YART, for interactively playing around with RCODE, I suggest >getting a WiFi setup and using 'telnet' to type in RCODE/RCodePlus commands >directly. |
| V 1510 Program conversion os3984 [2004/11/16-12:50] |
| | Response : Email : Edit : Delete |
|
| I am
trying to convert some OPEN-R programs developed initially for an
ERS-220A | in order to be used for the new ERS-7. I already changed locator definitions, | GAIN values, angles, etc. I can get information from AIBO-7 remotely | (debugging) and I know that it is working well (perfect). But I can not get | any movement (legs and head) from AIBO-7. Currently, I am using a memory | stick of 8MB. Should I use a memory stick of 16 MB? Anyone has some idea | about what is happening? | |
| +- 1512 RE:Program conversion AiboPet [2004/11/17-03:14] |
| | Response : Edit : Delete |
|
| >
Currently, I am using a memory stick of 8MB. Should I use a memory
stick of | 16 MB? | Shouldn't make any difference - as long as the program fits. | | >But I can not get any movement (legs and head) from AIBO-7. | > Anyone has some idea about what is happening? | | There are plenty of possibilities. | | Are you sure the routines moving the legs are being called ? | (SendData going to the OVirtualRobotComm.Effector) | | Common problems: | ? the connection in CONNECT.CFG got lost | ? the stretch case was not handled properly. If you stole the stretching code | from the samples, it should handle the cases properly. Does your DoStart | routine check "IsReady()" on the Effector? | |
| +- 1514 RE:Program conversion wangy22 [2004/11/17-17:06] |
| | Response : Edit : Delete |
|
| What
do you mean by strength case? | | I have the same problem. The program runs no problem on ERS210A, and after I did | the same thing as os3984, it does not work properly on ERS7: | | After the booting of AIBO, the legs just quiver for less than 1 second and then | stop there. | | Now only joints in head is moving, as well as the video/audio streaming. | | Yu | >> Currently, I am using a memory stick of 8MB. Should I use a memory stick of | >16 MB? | >Shouldn't make any difference - as long as the program fits. | > | >>But I can not get any movement (legs and head) from AIBO-7. | >> Anyone has some idea about what is happening? | > | >There are plenty of possibilities. | > | >Are you sure the routines moving the legs are being called ? | >(SendData going to the OVirtualRobotComm.Effector) | > | >Common problems: | >? the connection in CONNECT.CFG got lost | >? the stretch case was not handled properly. If you stole the stretching code | >from the samples, it should handle the cases properly. Does your DoStart | >routine check "IsReady()" on the Effector? | |
| +- 1517 RE:Program conversion AiboPet [2004/11/18-06:00] |
| Response : Edit : Delete |
|
>What
do you mean by strength case? The STRETCH case Take a look at the MovingLegs sample (2x0 or ERS7 version) There are several important things for normal movement: * a subject to send move commands to (stub.cfg) * in CONNECT.CFG a connection to OVirtualRobotComm.Effector * places in the code that SetData with joint movement commands For the startup stretch case, usually in your DoStart method, you need: * a routine to turn on the legs and get them ready to move ("AdjustDiffJointValues") - called in TWO places * when that is done, move the legs into idle position (eg: laying down) * to check to see if the effector is ready - using "IsReady", and call the stretch routine now or later (ie. why AdjustDiffJointValues is called in TWO places) If your code is radically different, then I suggest you incorporate those bits from the samples. |
| V 1513 Bad reboot ejt [2004/11/17-08:04] |
| Response : HomePage : Email : Edit : Delete |
|
Just
wondering if anyone else has had trouble with the AIBO (210 in
particular, perhaps 7 as well?) having its joints snap to extreme positions after a reboot. Note reboot - never happens after a full shutdown and startup power cycle I'm refering to an explicit reboot, such as from the FTP server after a 'REBT' command (which is where we are seeing the problem). * It only occurs occassionally, without an apparent pattern * A full power cycle clears the problem, so it's not bad data on the memstick I'm thinking there may be a problem with some memory not being properly initialized by the system after a reboot (reboots happen faster than full boots, indicating there's some shortcuts) -ethan |
| V 1495 Can we use OPEN-R SDK to create our own commercial products? Mustapha [2004/11/11-06:17] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hello! | | I'm CEO of software company. The thing I want to ask is: can we create our | custom commercial software using OPEN-R SDK for AIBO for distributing and | selling purposes (for example, as Shareware)? | | Thank you. | |
| +- 1496 RE:Can we use OPEN-R SDK to create our own commercial produc openrsupport@(Administrator) [2004/11/11-10:39] |
| | Response : Edit : Delete |
|
| I'm
sorry, but the OPEN-R SDK is for non-commercial use only. However, | commercial use is allowed if you use the R-CODE SDK, AIBO Remote Framework, or | AIBO Motion Editor. Please see the following FAQ: | | https://openr.aibo.com/openr/eng/perm/regi_faq.php4#A0026 | | AIBO SDE Support | |
| |- 1500 RE:Can we use OPEN-R SDK to create our own commercial produc ejt [2004/11/12-04:32] |
| | Response : Email : Edit : Delete |
|
| >I'm
sorry, but the OPEN-R SDK is for non-commercial use only. However, | >commercial use is allowed if you use the R-CODE SDK, AIBO Remote Framework, or | >AIBO Motion Editor. Please see the following FAQ: | | I've known for a while now that the OPEN-R SDK is not for commercial use, but | I'm surprised (and pleased) to hear you are approving these higher level tools | for commercial use. | | I get the feeling you want to be like Apple, and retain control over both | hardware and software. This has some advantages, but Apple further realizes | they wouldn't get any sales if they were the only ones who could sell software | for their platform - you'll sell a lot more hardware if there's a vibrant market | of software to run on it, and then you'll get more of your own software sales too. | | I think things have definitely moving in the right direction, first with the | opening of the OPEN-R SDK, and now the commercial use of the higher level | development tools - perhaps it's time to re-evaluate this policy now? | | Just a thought... | | thanks, | -ethan | |
| +- 1504 RE:Can we use OPEN-R SDK to create our own commercial produc wangy22 [2004/11/13-07:02] |
| Response : Edit : Delete |
|
In
this case, I can release JUST THE CODE I wrote for AIBO using OPEN-R libraries under GPL license? Thanks >I'm sorry, but the OPEN-R SDK is for non-commercial use only. However, >commercial use is allowed if you use the R-CODE SDK, AIBO Remote Framework, or >AIBO Motion Editor. Please see the following FAQ: > >https://openr.aibo.com/openr/eng/perm/regi_faq.php4#A0026 > >AIBO SDE Support |
| V 1502 How can I use OPEN-R sdk to ON/OFF for each face LED of ERS7 samlu [2004/11/12-11:36] |
| | Response : Email : Edit : Delete |
|
| I'd like to use OPEN-R SDK to write a problem to show | a analog clock on ERS-7's face. | | After a few studing and tring, it seems that I can only | show the predefined LED patterns using modeA or modeB method. | | Are there other methods(or APIs) for me to control the face LEDs one by one? | |
| +- 1503 RE:How can I use OPEN-R sdk to ON/OFF for each face LED of ERS7 AiboPet [2004/11/13-02:13] |
| Response : Edit : Delete |
|
>Are
there other methods(or APIs) for me to control the face LEDs one by one? To reduce cost and complexity (especially for those expensive white LEDs), the holes in the ERS-7 faceplate, are just that, holes. The actual controllable LEDs are behind the scenes, and they are much fewer than the actual holes (ie. one "LED" lights up several holes) I suggest running Skitter (ERS7 LED editor) if you want to play around with the different combinations. |
| V 1499 Technical chat on aibo-life.org tonight openrsupport@(Administrator) [2004/11/12-02:57] |
| Response : Edit : Delete |
|
Hi, A Sony representative will be present at an unofficial technical chat at the third-party website www.aibo-life.org tonight (Thursday) at 6 PM Pacific time / 9 PM Eastern time / 2 AM GMT. This is NOT an official AIBO SDE event. However, since it is a technical chat, you may be interested and are very welcome to join. Thanks, AIBO SDE Support |
| V 1494 Request: minimal RFW info to allow CLIE/PocketPC apps AiboPet [2004/11/11-03:43] |
| | Response : Edit : Delete |
|
| Now
that RFW is final, it would be nice to remove the reliance on the PC
DLLs | so that RFW apps could be written for the PalmOS (especially for Sony CLIE) | and/or PocketPC devices. | | Most of the socket protocols appear to be straight forward, including the | minimal documentation of which sockets are used. | However the actual socket data formats are not documented (in particular the | connection/login password security mechanism) | | My request: | Provide enough documentation or source code for others (ie. me) to create | PalmOS and/or PocketPC bindings. | If this requires disabling the login password security mechanism (so you don't | have to document the security protocols), that's fine with me. | |
| +- 1497 RE:Request: minimal RFW info to allow CLIE/PocketPC apps openrsupport@(Administrator) [2004/11/11-10:58] |
| Response : Edit : Delete |
|
Thank
you for sharing your interest. Unfortunately, at this time we will not
be able to provide any more information about the protocol between AIBO and PC. There is a possibility that this protocol will be changed. If the protocol is changed, only VAIBOServer.exe and *.dll files will be affected. Your source code would remain the same. AIBO SDE Support |
| V 1470 Serial Bus bpaul [2004/11/02-00:52] |
| | Response : Edit : Delete |
|
| In
Aibo each module is connected by a high speed serial bus, does anyone
has | more information on this bus. | Moreover Aibo's configuartion is extensible, So i was wondering if connecting it | with a new sensor would be an possibility or not ? | |
| +- 1473 RE:Serial Bus AiboPet [2004/11/05-00:03] |
| | Response : Edit : Delete |
|
| >In
Aibo each module is connected by a high speed serial bus, does anyone
has | >more information on this bus. | Nothing has been published recently. | There are some very old ACM articles that talk about the "OpenR Serial Bus" | back in the days before AIBO - when it was a "Mutant" robot. | The serial bus appears to be very similar to USB 1.x (12Mbps) and I believe | the protocols are similar too [the ERS-11x models have a place for a USB | connector too] | | >Moreover Aibo's configuartion is extensible, So i was wondering if connecting | it | >with a new sensor would be an possibility or not ? | It would be nice if they published the specs, but I'm not holding my breath. | | I've considered adding sensors, but IMHO AIBO already has all the important | built-in sensors. | Alternatives: In some cases I've strapped on other devices (eg: communicating | with a PC with BT/WiFi). Also it is possible to hack a robot to access the | serial port connections. | | What kind of sensors / addons are you interested in ? | |
| +- 1475 RE:Serial Bus bpaul [2004/11/05-00:15] |
| | Response : HomePage : Edit : Delete |
|
| I
was thinking about a fire sensor. | Moreover,this additional new "component" should be defined/implemented as a | software component in the underlying Aperios OS. | | The a new Sytem Object has to be defined say VirtualRobotFireCommunicator | or maybe under the OVirtualRobotCommunication with a corresonding outgoing Gate | lets say "Fire". | | It all sound so attractive, but next to impossible. | |
| +- 1476 RE:Serial Bus AiboPet [2004/11/06-02:06] |
| | Response : Edit : Delete |
|
| >I
was thinking about a fire sensor. ... | | For something that simple, I would suggest rewiring one of the existing switch | sensor that you don't otherwise need | [eg: the jaw sensor, or one of the paw sensors] | |
| +- 1477 RE:Serial Bus bpaul [2004/11/06-21:01] |
| | Response : Edit : Delete |
|
| and
what abou the software interface to that ? | Suppose i want ot put an radio antenna to trace its location. I would like to add it to the complete | model, not by removing one of its other switch sensors. | |
| +- 1481 RE:Serial Bus AiboPet [2004/11/08-01:03] |
| | Response : Edit : Delete |
|
| >and
what abou the software interface to that ? | ... | The existing system interface for reading the switch would work. | I was suggesting an alternative for simple (binary) inputs. | | For more elaborate things, I suggest hooking up a small WiFi device (eg: PDA) | and have it do the extra interfacing. AIBO can talk to it with WiFi. Other | alternatives are possible if you have a PC in the equation. If willing to take | apart your robot and perform hardware mods, there are other possibilities... | | [ie. instead of waiting for the OpenR Serial Bus to be opened up by Sony...] | |
| +- 1492 RE:Serial Bus bpaul [2004/11/10-19:32] |
| | Response : Edit : Delete |
|
| Thinking
on the same lines, I am sure it is impossible to install another
(better) camera in Aibo in place | of its existing one because the DSP takes place in the CPU. | | One can never extend the posibilties of what the "System Core" can handle.. so to speak. | > | >For more elaborate things, I suggest hooking up a small WiFi device (eg: PDA) | >and have it do the extra interfacing. AIBO can talk to it with WiFi. Other | >alternatives are possible if you have a PC in the equation. If willing to take | >apart your robot and perform hardware mods, there are other possibilities... | | Can u elaborate more on this ? | Both on WiFi possibilties and hardware mods. | |
| +- 1493 RE:Serial Bus AiboPet [2004/11/11-02:59] |
| Response : Edit : Delete |
|
FWIW:
I'd like to see an official response to your original question (ie. any chance in heck that they will document the serial protocol) ---- UNOFFICIAL ANSWERS: >I am sure it is impossible to install another (better) camera in Aibo... True, there is much custom hardware to do the color filtering. >>For more elaborate things, I suggest hooking up a small WiFi device (eg: PDA) >>and have it do the extra interfacing. >Can u elaborate more on this ? For example, take a WiFi enable PDA (eg: certain Sony CLIE devices, or a PocketPC) and interface it to your custom hardware [that's a topic on its own]. Then the WiFi PDA can send information to the AIBO (both talking WiFi). You can organize it so the PDA is in control or the AIBO is in control - up to you. Alternatively, you can coordinate this through a WiFi PC (allowing other options like programming with the RFW, or if you are tricky using both BlueTooth and WiFi) The disadvantage is you need to strap the PDA and other electronics to AIBO's back (adding weight) >>taking apart your robot and perform hardware mods, there are other possibilities... I've figured out how to hook up the serial port for simple debugging output, but it is not an ideal solution. If you are not opposed to taking apart the AIBO and experimental soldering, and adding a single chip CPU, many things are possible if you roll your own interface to the existing AIBO brain [eg: pulse the existing switch inputs with a coded signal] IMHO: such hardware mods are better with cheaper raw material (ICybie or RoboSapien robots) |
| V 1491 R-CODE for ERS-110/111 and R-CODE 1.2 for ERS-210/220/310 openrsupport@(Administrator) [2004/11/10-11:37] |
| Response : Edit : Delete |
|
R-CODE
for the ERS-110/111 and R-CODE 1.2 for the ERS-210/220/310 is now available for download on this website. The R-CODE archive can be found on the R-CODE download page. https://openr.aibo.com/openr/eng/perm/rcode/download/download.php4 Thank you, AIBO SDE Support (The article that 2004/11/10-11:36 posted, was modified by Author) |
| V 1490 AIBO Remote Framework (Final) has been released. openrsupport@(Administrator) [2004/11/10-09:59] |
| Response : Edit : Delete |
|
* AIBO Remote Framework (Final) has been released. AIBO Remote Framework (Final) is incorporated into AIBO MIND 2. Please refer to the FAQs to learn how to use it. [2004/11/10] https://openr.aibo.com/openr/eng/perm/rfw/download/download.php4 https://openr.aibo.com/openr/eng/perm/rfw/regi_faq.php4 AIBO Remote Framework beta cannot be downloaded any more. Please use AIBO Remote Framework (Final) with AIBO MIND2. Thank you, AIBO SDE Support |
| V 1488 making AIBO play a wav file benjaminvigier [2004/11/09-22:18] |
| | Response : Email : Edit : Delete |
|
| Hi, | i am having a hard time trying to make AIBO playback a 8K8bit wav file. I am | using AIBO RFW, the doc does not say much about it, there are a few methods | (playwavedata(), AudioSendData(), etc.) but no explanations attached. The | samples provided illustrate everything except this functionnality...well, in | other words, i am quite confused.. :) | If somebody could be kind enough to paste the piece of code that does it, i'll | offer him a beer the next time he comes by paris..:) | thanks, | benjamin | |
| +- 1489 RE:making AIBO play a wav file AiboPet [2004/11/10-01:33] |
| Response : Edit : Delete |
|
>Hi, >i am having a hard time trying to make AIBO playback a 8K8bit wav file.... If you can send the entire WAV file at one time, it is very easy: int err = pVAIBO->AudioSendWAVFile("whatever.wav"); if (err != 0) oops_something_went_wrong(...); ---- Before that you should have called AudioInit, AudioSendOpen (pass the right format to AudioSendOpen) and RequestNetService(NETAUDIO_ID) I don't know if all of these are absolutely required. |
| V 1471 Author DELETE ***** [2004/11/09-13:53] |
| Response : Edit : Delete |
|
(The
article that 2004/11/04-12:20 posted, was modified by Author) |
| V 1483 Limitations of J1 joints of ERS-7's left leg and right leg wangy22 [2004/11/08-10:40] |
| | Response : Edit : Delete |
|
| (Sorry for my bad English....) | | For ERS-210a, the software limitations is [-120, 120] for J1 joints of both | right leg and left leg. | | However, for ERS-7, this J1 limitation is separated into two: [-115, 130] for | left leg and [-130, 115] for right leg. | | Is it a typing error of the document? | | I hope it is, since I am now porting the program I've written for ERS-210a to | ERS-7, I found it will be inconvient in rewriting the module for leg joints... | | If it is not an error, any suggestions on my work? | | Thanks a lot! | |
| +- 1484 Problems also found in MovingLegs7.cc from the given samples wangy22 [2004/11/08-12:06] |
| | Response : Edit : Delete |
|
| I've
just read the sample program MovingLegs7.cc and found that: | | const double BROADBASE_ANGLE[] = { | 120, // RFLEG J1 | 90, // RFLEG J2 | 30, // RFLEG J3 | | 120, // LFLEG J1 | 90, // LFLEG J2 | 30, // LFLEG J3 | | -120, // RRLEG J1 | 70, // RRLEG J2 | 30, // RRLEG J3 | | -120, // LRLEG J1 | 70, // LRLEG J2 | 30 // LRLEG J3 | }; | | Since the lower bound for LRLEG J1 is -115 degrees, according to the Model | Information documents for ERS-7, how is that possible for LRLEG J1 to be -120 | degrees? | | Since the upper bound for RFLEG J1 is 115 degrees, according to the same Model | Information documents, how is it possible for RFLEG J1 to be 120 degrees? | | APPENDIX: Software Limitations of Joint Motions from Model Information for ERS7: | | 2.1.1 Single Joints Parts Range (Unit: degree) | Neck tilt1 (-75, 0) | Neck pan (-88, 88) | Neck tilt2 (-15, 45) | Mouth (-55, -3) | Left leg(front/rear)J1 (-115, 130) | Left leg(front/rear)J2 (-10, 88) | Left leg(front/rear)J3 (-25, 122) | Right leg(front/rear)J1 (-130, 115) | Right leg(front/rear)J2 (-10, 88) | Right leg(front/rear)J3 (-25, 122) | Tail tilt (5, 60) | Tail pan (-45, 45) | |
| +- 1487 RE:Problems also found in MovingLegs7.cc from the given samp openrsupport@(Administrator) [2004/11/09-10:50] |
| Response : Edit : Delete |
|
Thank
you for noticing the discrepancy. The MovingLegs7 source code is
correct, the ERS-7 Model Information document is wrong. The J1 joint limitations are: Front J1 (R/L) [-115, 130] Rear J1 (R/L) [115, -130] The ERS-7 Model Information document will be updated in the next release of the documentation. Thank you for your report. AIBO SDE Support |
| V 1467 Simple Open-R introduction nlopes [2004/10/30-02:07] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I'm new to Open-R and I'm finding difficult to find some documentation about | it. | I would like to know how can I do simple things like read a joint value, | modify a sensor/joint value, etc... | | Thanks, | Nuno | |
| |- 1468 RE:Simple Open-R introduction nlopes [2004/10/31-22:05] |
| | | Response : Edit : Delete |
|
| | And
why do you need OPENR::GetJointValue() ? | | | | It doesn't work. You need to connect to OVirtualRobotCommSensor in order to | | get the joint values... | | |
| | +- 1469 Author DELETE ***** [2004/11/01-09:36] |
| | Response : Edit : Delete |
|
| (The
article that 2004/11/01-09:36 posted, was modified by Author) | |
| |- 1474 RE:Simple Open-R introduction AiboPet [2004/11/05-00:12] |
| | Response : Edit : Delete |
|
| >I'm
new to Open-R and I'm finding difficult to find some documentation about | >it. | | Join the club ;-> | The documentation can be minimal at times. | | I assume you have read through the included documentation. | You also should spend some time reading through the sample code, at least for | the basic samples (HelloWorld, ObjectComm, MovingHead/Legs) | | There is also a 3rd party tutorial: | http://www.ensta.fr/uer/uei/baillie/eng/openr_tutorial.html | | >I would like to know how can I do simple things like read a joint value, | >modify a sensor/joint value, etc... | | The samples do that. | | NOTE: reading a joint value is not a common operation. Usually you tell the | robot what angle the joints should be and the robot will move them to that | position. As in the samples, the only time you should call | OPENR::GetJointValue is in the initial "stretching" code (which I suggest | stealing from one of the samples) | |
| +- 1479 RE:Simple Open-R introduction pyrol [2004/11/08-00:23] |
| Response : Edit : Delete |
|
FYI: "OPEN-R SDK School" on this website may help. Go to the "AIBO SDE University" page by clicking the "University" button. Then click the link "(2-1) OPEN-R SDK School". |
| V 1472 UDP aibourjc [2004/11/04-21:21] |
| Response : HomePage : Email : Edit : Delete |
|
I
am changing this programm so that it sends bigger packets. I change this line in #define UDPECHOSERVER_BUFFER_SIZE 32000 in "UDPEchoServerConfig.h". the maximun packet size is supposed to be 64kbytes, and in this example the createMsg function is called : udpCreateMsg(EndpointType_UDP,UDPECHOSERVER_BUFFER_SIZE * 2); but then, the maximun size I can really send to the PC is a 19Kbyte message, with bigger sizes, or when sending for example a 18'5 Kbytes several times one after the other, I get a crash, and when reading the information of emon.log it says it is in IPStack module. when the defined size is for example of 12k, if I send bigger packages, I get chrases, but when reaching the 19kb packet size, the definition of UDPECHOSERVER_BUFFER_SIZE does not seem to have any effect. Also, I find it strange that the program chrases instead of giving and ejecution error in one of the OPENR functions. Can anybody help me? Thank you so much!!! |
| V 1464 MS Visual Studio 2003 / OPEN-R SDK bradleyjs [2004/10/28-05:15] |
| | Response : Email : Edit : Delete |
|
| Haven't
started to write/compile anything yet (as I just got my memory stick | R/W device), but has anyone had any issues compiling with Microsoft's Visual | Studio 2003? | | Just curious. If so, what is the recommended compiler for a WinXP box? | |
| +- 1465 RE:MS Visual Studio 2003 / OPEN-R SDK bradleyjs [2004/10/28-05:20] |
| | Response : Email : Edit : Delete |
|
| ***Download
Cygwin, MIPS cross-development tools*** | Now that I've read the FAQ, I guess this is the tool of choice? | | I will certainly give MSVS 2003 a try and see what happens...I use it almost | every day and am very used to it's interface... | | >Haven't started to write/compile anything yet (as I just got my memory stick | >R/W device), but has anyone had any issues compiling with Microsoft's Visual | >Studio 2003? | > | >Just curious. If so, what is the recommended compiler for a WinXP box? | |
| +- 1466 RE:MS Visual Studio 2003 / OPEN-R SDK bradleyjs [2004/10/28-05:58] |
| Response : Edit : Delete |
|
Yep,
looks like Cygwin is the only way---library files aren't compatible... >***Download Cygwin, MIPS cross-development tools*** >Now that I've read the FAQ, I guess this is the tool of choice? > >I will certainly give MSVS 2003 a try and see what happens...I use it almost >every day and am very used to it's interface... > >>Haven't started to write/compile anything yet (as I just got my memory stick >>R/W device), but has anyone had any issues compiling with Microsoft's Visual >>Studio 2003? >> >>Just curious. If so, what is the recommended compiler for a WinXP box? |
| V 1460 ERS-210 Programmable Memory Sticks (PMS) 8MB or 16MB bradleyjs [2004/10/27-22:01] |
| | Response : Email : Edit : Delete |
|
| Does
anyone know if the 16MB (ERA-MS016) PMS will work with an original ERS- | 210? I can't find any 8MB (ERA-MS008) available for purchase only the 16MB | PMS's. | |
| +- 1461 RE:ERS-210 Programmable Memory Sticks (PMS) 8MB or 16MB AiboPet [2004/10/28-00:43] |
| | Response : Edit : Delete |
|
| >Does
anyone know if the 16MB (ERA-MS016) PMS will work with an original ERS- | >210? | | Yes, the 16MB PMS sticks will work in any model AIBO | $39 USD, available at SonyStyle or RoboToys | | The older 8MB PMS sticks were discontinued in the US and Europe a long time | ago. | |
| +- 1463 RE:ERS-210 Programmable Memory Sticks (PMS) 8MB or 16MB bradleyjs [2004/10/28-02:10] |
| Response : Email : Edit : Delete |
|
Thanks
for the info, I figured as such. I will order them from SonyStyle today. >>Does anyone know if the 16MB (ERA-MS016) PMS will work with an original ERS- >>210? > >Yes, the 16MB PMS sticks will work in any model AIBO >$39 USD, available at SonyStyle or RoboToys > >The older 8MB PMS sticks were discontinued in the US and Europe a long time >ago. |
| V 1459 I want to know the version history of OPEN-R ptkang [2004/10/27-19:42] |
| | Response : Email : Edit : Delete |
|
| I wannt to know the version history of OPEN-R Project. | When did this project start and What is the issue of each version? | |
| +- 1462 RE:I want to know the version history of OPEN-R AiboPet [2004/10/28-00:59] |
| Response : Edit : Delete |
|
>I
wannt to know the version history of OPEN-R Project. >When did this project start and What is the issue of each version? Do you mean "Open-R" or the "Open-R SDK" ? ----- re: the general term "Open-R" I believe the term "Open-R" was first announced publicly in 1998. Press release from 1998: http://www.sony.net/SonyInfo/News/Press/199806/98- 052/index.html There are earlier research papers (before 1998) as well. Search this post for any live links: http://www.robots.net/article/81.html Earlier versions were for research only. OpenR is built upon a Sony proprietary operating system called Aperios (earlier called Apertos) which goes much further back. The first AIBO (ERS-110) which uses OpenR was released in mid 1999. The ERS-210 generation of OpenR software was released in late 2000. The current generations of AIBOs (210, 220, 31x and ERS-7) are evolved from that version. Up until mid 2002, "OpenR" was a Sony proprietary interface. RoboCup participants had to sign an NDA to get access to the RoboCup version of OpenR. --- The "Open-R SDK" is a public SDK, that went online in mid 2002. That's what can be downloaded and is discussed on this BBS. (ie. up until then, Open-R was closed and proprietary) There have been several updates of this (originally for the 210/220, current version for the ERS-7 too) Check the "What's New" box for some of the history of updates - ie. relatively minor IMHO compared to the evolution before 2002. Just this year, the "Open-R SDK" started including other tools, and a new term "SDE" was invented (covers the Open-R SDK and much more) |
| V 1457 LED and Behavior cale466 [2004/10/25-19:09] |
| | Response : Email : Edit : Delete |
|
| I'm
conducting a research project on the emotional features of AIBO ERS-7.
Is | there any way to find out all the possible LEDS and what they represent? | Also, what kind of behaviors it displays depending on the five instincts and | six emotions? | Thanks in advance! | |
| +- 1458 RE:LED and Behavior AiboPet [2004/10/26-00:13] |
| Response : Edit : Delete |
|
>
Is there any way to find out all the possible LEDS and what they
represent? The OpenR SDK has a technical discussion of the face LED array. Rather dry reading IMHO. I recommend the Skitter LED editor. You can play around with it and try different combinations. http://www.dogsbodynet.com/skitter.html ======= >Also, what kind of behaviors it displays depending on the five instincts and >six emotions? There are no stead-fast rules. In general: Green for happy/joyful Red for angry Purple for sad or confused The white LEDs allow for interesting combinations. It can show two eyes, one eye (wink), a Bone shape, small eyes (fearful), various cross shapes or limited changing images/animations. The brightness of the white LEDs also can mean something Dim crosses can mean he is demur (like small eyes) Bright crosses can mean surprise Large crosses with red means he is pissed off ---- For a thorough analysis you need access to the system .LED files - which unfortunately are encrypted. Similar conclusions can be drawn by playing around with YART7, and making AIBO perform actions with moods. http://www.aibohack.com/ers7/yart7.htm The canned action names are derived from the internal storage names, many of which include mood as part of the name. eg: "L_REFLEX.M_SAD.S_SOUND.6" is one sadness action (there are a lot of them) Some guesswork as to how the names are used in the regular AiboMind personality is needed. |
| V 1448 How to port ERS210 Program to ERS7 wangy22 [2004/10/16-07:26] |
| | Response : Edit : Delete |
|
| I
have developed a program for ERS210 in last summer and I am getting a
ERS-7 | now. So I want to know, in minimum, what should be modified to port the existing | program to ERS-7? Are modifications on just primitiveID and hardware limits enough? | | Thanks a lot! | |
| +- 1449 RE:How to port ERS210 Program to ERS7 TodboT [2004/10/18-11:27] |
| | Response : Edit : Delete |
|
| You'll
have to change the hardware locators, and the joints aren't exactly the | same (neck). You'll need to use the memory stick image for the 7, but other | than that, I don't think you really need to do anything else. | |
| +- 1456 RE:How to port ERS210 Program to ERS7 AiboPet [2004/10/25-03:11] |
| Response : Edit : Delete |
|
Also: The camera resolution is different. If you are using LEDs, they are all different. You may want to use the new IR distance sensor in the body, as well as the new back sensors. |
| V 1452 Is there any hardware difference between ERS-7 and ERS-7M2? marsdog [2004/10/20-16:53] |
| | Response : Email : Edit : Delete |
|
| We
are planning to buy some new Aibos for RoboCup. The new ERS-7M2 will be | available next month but we can not decide whether there is any difference | between ERS-7M2 and ERS-7 only from the introductions on the web. | I guess ERS-7M2 means ERS-7 Mind2, so it's only a new version with updated | software but the same hardware, isn't it? | Thanks. | |
| +- 1453 RE:Is there any hardware difference between ERS-7 and ERS-7M2? slider2nl [2004/10/21-09:37] |
| | Response : Email : Edit : Delete |
|
| both
aibo's have the same hardware only the color is diffrent. | | the m2 is aibomind2 | | >We are planning to buy some new Aibos for RoboCup. The new ERS-7M2 will be | >available next month but we can not decide whether there is any difference | >between ERS-7M2 and ERS-7 only from the introductions on the web. | >I guess ERS-7M2 means ERS-7 Mind2, so it's only a new version with updated | >software but the same hardware, isn't it? | >Thanks. | (The article that 2004/10/21-09:36 posted, was modified by Author) | |
| +- 1455 RE:Is there any hardware difference between ERS-7 and ERS-7M2? AiboPet [2004/10/25-03:09] |
| Response : Edit : Delete |
|
>both
aibo's have the same hardware only the color is diffrent. And the charging station is different (black and white target, not blue and green). If you purchase the AiboMind2 upgrade, it comes with the new charging station parts. The robots themselves are the same (however the first batch of ERS-7s from last year have possible joint problems -- see the free service offer). Of concern if you are buying a second-hand ERS-7 |
| V 1454 Announcing Tekkotsu 2.2 ejt [2004/10/21-10:38] |
| Response : HomePage : Email : Edit : Delete |
|
Tekkotsu version 2.2 has been released! http://www.tekkotsu.org/ We also have a new introductory Tutorial under development by Dr. Touretzky: http://www-2.cs.cmu.edu/~dst/Tekkotsu/Tutorial/ ---------------------------------------------------------------------------- New features include: .. * New kinematics library: http://www.tekkotsu.org/Kinematics.html .. * Path specification and execution via waypoint list ...... http://www-2.cs.cmu.edu/~tekkotsu/dox/classWaypointEngine.html .. * UDP support for smoother and more robust video streaming ...... (thanks Bryan Johnson and Erik Berglund) .. * Sound input support (thanks Paris Smaragdis) .. * EventRouter status events, notify generators of added/removed listeners .. * The ERS-7 model is now the default build target. .. * More flexible build system, switch between targets without recompile ...... - See the configuration setup for instructions on switching targets: .......... http://tekkotsu.no-ip.org/TekkotsuInstall.html .. * and more! http://www-2.cs.cmu.edu/~tekkotsu/VersionHistory.html#v2.2 ---------------------------------------------------------------------------- The new Kinematics library represents a large body of work, and we hope it will prove useful to anyone doing manipulations. We have done a significant amount of painful (for me anyway) math to allow you to solve for the joint angles needed to put any point on any link to any point in space, or as close as it can get. Many thanks to everyone who has provided patches! We look forward to seeing the exciting stuff you are working on! |
| V 1423 ERS-7 Joint speed limits? ejt [2004/10/06-07:05] |
| | Response : HomePage : Email : Edit : Delete |
|
| Once
upon a time I obtained recommended speed limits for the joints of the | ERS-2xx series, but I can't remember where. | | Would the administrators like to provide a set of recommended speed limits for | the joints of the ERS-7? | | thanks! | -ethan | |
| |- 1425 RE:ERS-7 Joint speed limits? bpaul [2004/10/06-18:28] |
| | | Response : Edit : Delete |
|
| | My qustion is in the same direction: | | | | Aibo ERS-110 had a angular velocity sensor(torso) but where is it present in ERS-7 | | I would like to claculate the maximum limit of anular vel. to implement fat motion. | | |
| | +- 1428 RE:ERS-7 Joint speed limits? ejt [2004/10/07-06:53] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | >My
qustion is in the same direction: | | > | | >Aibo ERS-110 had a angular velocity sensor(torso) but where is it present in ERS-7 | | >I would like to claculate the maximum limit of anular vel. to implement fat motion. | | | | The ERS-7 (and 2xx) has 3 accelerometers (x,y,z) but no angular velocity sensor | | for the motion of the body itself, if that's what you mean. | | | | The limits I'm requesting are the recommended movement guidelines to prevent | | excessive wear, which will be well below the maximum possible joint speeds, | | which it sounds like is what you want. | | This would probably be best described by giving the motor torques and link | | inertia matricies, which I would love to see as well, but to my knowledge have | | never been publicly provided. :( | | (The article that 2004/10/07-06:51 posted, was modified by Author) | | |
| | +- 1436 RE:ERS-7 Joint speed limits? bpaul [2004/10/12-19:18] |
| | | Response : HomePage : Edit : Delete |
|
| | So
where are these: "recommended movement guidelines" ? Perhaps that would
give | | an idea. | | >The limits I'm requesting are the recommended movement guidelines to prevent | | >excessive wear, which will be well below the maximum possible joint speeds, | | >which it sounds like is what you want. | | | | I was wondering how Motion Editor calculates which interpolation sequence of | | action (from one posture to another) is safe or not ? | | |
| | +- 1441 RE:ERS-7 Joint speed limits? ejt [2004/10/13-11:27] |
| | Response : HomePage : Email : Edit : Delete |
|
| >So
where are these: "recommended movement guidelines" ? Perhaps that would
give | >an idea. | That's the whole point ;) I'm asking for them to be posted. | | >I was wondering how Motion Editor calculates which interpolation sequence of | >action (from one posture to another) is safe or not ? | Yes, by definition, actually... the limits I'm looking for are those used by Motion Editor to limit motion | speed. I think, anyway. I was hoping someone might have the table handy, but I guess I can try to | figure it out from within Motion Editor. | |
| +- 1440 RE:ERS-7 Joint speed limits? TodboT [2004/10/13-09:44] |
| | Response : Edit : Delete |
|
| This
isn't exactly what you're looking for, but you can use MEdit and "check" | the motion to see if it's safe. | (The article that 2004/10/13-09:43 posted, was modified by Author) | |
| +- 1442 RE:ERS-7 Joint speed limits? ejt [2004/10/13-12:24] |
| | Response : HomePage : Email : Edit : Delete |
|
| >This
isn't exactly what you're looking for, but you can use MEdit and "check" | >the motion to see if it's safe. | Yeah, unfortunately I want to use this on the OPEN-R SDK side of the fence, with dynamic motions, so | it's a more general problem than just checking one motion. | Perhaps I'll eventually get around to figuring out the limits MEdit uses, but I was hoping someone might | have had this information on hand... *shrug* | |
| +- 1450 RE:ERS-7 Joint speed limits? openrsupport@(Administrator) [2004/10/18-17:51] |
| | Response : Edit : Delete |
|
| MEdit
uses the following angular speed limits: | | Tilt1 = 1.46 (degree/frame) | Pan = 4.61 (degree/frame) | Tilt2 = 2.65 (degree/frame) | Mouth = 4.65 (degree/frame) | Leg J1 = 2.23 (degree/frame) | Leg J2, J3 = 2.42 (degree/frame) | Tail Pan,Tilt = 6.95 (degree/frame) | | 1 frame = 8 msec | | However, the angular acceleration limits are more important than the angular | velocity limits. | | Please use slower starting and stopping motions to protect AIBO's motors. | | PID parameters control the angular accelerations. | If you are using the OPEN-R SDK, use the recommended PID parameters in the MoNet | sample. In the R-CODE SDK and AIBO Remote Framework, PID values are | automatically controlled in the middle ware object for each built-in motion. | Different PID parameters are used in each motion to control AIBO more safely. | | The angular velocity limits of MEdit are based on the | R-CODE SDK and AIBO Remote Framework. | | Thank you, | | AIBO SDE Support | |
| +- 1451 Thanks! (nt) ejt [2004/10/19-05:22] |
| Response : HomePage : Email : Edit : Delete |
|
|
| V 1433 AIBO ERS_7 Iaia [2004/10/11-20:00] |
| | Response : Email : Edit : Delete |
|
| Hallo!I'm Nadia, | I study Computer Science at the University of Bari in Italy. | AIBO ERS-7 will arrive in faculty in few day. I have decided to prepare the | thesis on AIBO. Can I have an advice on SDK and the Tools to be used to | programm its? What could I make its do? | Thank you in advance. | Nadia | (The article that 2004/10/11-19:59 posted, was modified by Author) | |
| +- 1439 RE:AIBO ERS_7 AiboPet [2004/10/13-01:07] |
| | Response : Edit : Delete |
|
| >AIBO
ERS-7 will arrive in faculty in few day. I have decided to prepare the | >thesis on AIBO. | | Do you have a topic ? | | > Can I have an advice on SDK and the Tools to be used to | >programm its? What could I make its do? | | Many things are possible. | | First, I suggest playing with the dog when you get it, change it to adult and | see how it moves when used normally. As a regular user, no programming. | | My general advice for all programming is to start with YART | YART for the ERS-7 http://www.aibohack.com/ers7/yart7.htm | It lets you program custom behavior very easily. | | There are several more advanced tools, with increasing complexity: | Straight RCODE programming, RFW, OpenR SDK and variants. | Depending on the kinds of technical problems you want to solve (based in part | on what you decide for your thesis) one tool may be better than the others. | |
| +- 1443 RE:AIBO ERS_7 Iaia [2004/10/14-01:16] |
| | Response : Email : Edit : Delete |
|
| My
thesis's topic is domotica and affective computing. For the moment my | lecturer has asked me to create an interface that shows than AIBO sees.Of what | tool I need? In Lab we have an access point and so I can check AIBO from my | computer. To programm AIBO I thought to use RFW and the Editor Motion why they | have suitably done for the Wireless connection.Is it right? | Thank you in advance. | Nadia | |
| +- 1447 RE:AIBO ERS_7 AiboPet [2004/10/15-06:54] |
| Response : Edit : Delete |
|
>
For the moment my >lecturer has asked me to create an interface that shows than AIBO sees.Of what >tool I need? Getting the camera image is the easiest case. There are several ways of doing this for the ERS-7: + AScope7 [runs on an enhanced version of AiboMind1] http://www.aibohack.com/ers7/scope7.htm + AiboSpeedRacer [camera data and walking algorithm] http://www.aibohack.com/openr_sdk/speedracer.htm + RFW (SimpleNavi and RemoteTest) + Tekkotsu - image viewer / Java UI controls etc. ---- > To programm AIBO I thought to use RFW and the Editor Motion why they >have suitably done for the Wireless connection.Is it right? I agree. I strongly recommend the RFW (AIBO Remote Framework) for those kinds of applications. It provides the basics like image capture, standard walking as well as the most advanced built-in primitives (like voice recognition, human face detection and vision pattern recognition) |
| V 1434 facial recognition and Learning on the Aibo mexico2003 [2004/10/11-23:42] |
| | Response : Edit : Delete |
|
| Hello there | | I was wondering if the following is possible in the Aibo dog. | | I want the dog to see a human face (with black hair) and walk towards it, when | it gets to the human, the human would pat it on the head sensor. This pat | would tell the dog that this person is a good person. | | Another human face (with blonde hair) would appear and the same would happen | but no pat would occur, this would tell the dog this is a bad person. | | Over time, I hope that the dog would be able to learn that walking to black | hair people would get it a pat and that the dog would not walk towards people | with blonde hair. | | Is this idea programmable with the dog and if so is there anywhere I can get | information about this. | | cheers | |
| +- 1438 RE:facial recognition and Learning on the Aibo AiboPet [2004/10/13-01:02] |
| | Response : Edit : Delete |
|
| >Hello there | > | >I was wondering if the following is possible in the Aibo dog.... | | Technically many things are possible... | | If you want to write your own human face detection algorithm, you can. | Aibo already has a human face detection module (available in RFW, but not | OpenR or standard RCODE). | The built-in human face detector may not suit your needs, since it usually | requires a close-up photo of a face, but it is a good place to start. | | I would suggest looking at RFW for a number of other reasons as well | Requires an ERS-7 and a full time WiFi connection to a Windows PC | | With RFW you can supplement it with additional code (ie. quickly process the | image to decide hair color, store additional logic on the PC etc) written in | C++ or almost any other language. | | RFW has the most built-in features, but no ability of adding your own programs | on the dog itself (programs must run on the PC, and control the dog remotely) | |
| |- 1444 RE:facial recognition and Learning on the Aibo Fairlawn07 [2004/10/14-04:46] |
| | Response : Edit : Delete |
|
| >>Hello
there | >> | >>I was wondering if the following is possible in the Aibo dog.... | > | >Technically many things are possible... | > | >If you want to write your own human face detection algorithm, you can. | >Aibo already has a human face detection module (available in RFW, but not | >OpenR or standard RCODE). | >The built-in human face detector may not suit your needs, since it usually | >requires a close-up photo of a face, but it is a good place to start. | > | >I would suggest looking at RFW for a number of other reasons as well | >Requires an ERS-7 and a full time WiFi connection to a Windows PC | > | >With RFW you can supplement it with additional code (ie. quickly process the | >image to decide hair color, store additional logic on the PC etc) written in | >C++ or almost any other language. | > | >RFW has the most built-in features, but no ability of adding your own programs | >on the dog itself (programs must run on the PC, and control the dog remotely) | | I have a question to AiboPet. Could you elaborate more about the human | detection algorithm in RFW? Also, maybe you could point me in a direction where | I can find more info about RFW? | |
| +- 1445 RE:facial recognition and Learning on the Aibo Fairlawn07 [2004/10/14-04:53] |
| | Response : Edit : Delete |
|
| >>Hello
there | >> | >> | >>Aibo already has a human face detection module (available in RFW, but not | >>OpenR or standard RCODE). | | >I have a question to AiboPet. Could you elaborate more about the human | detection algorithm in RFW? How do I access it? Also, maybe you could point me | in a direction where I can find more info about RFW? | | > | >) | |
| +- 1446 RE:facial recognition and Learning on the Aibo AiboPet [2004/10/15-06:47] |
| Response : Edit : Delete |
|
>
where I can find more info about RFW? Right here on this website. Press the "AIBO Remote Framework" button at the top of the page. The current Downloads are for a beta. The final version of RFW will be built-in to the AiboMind2 software that is coming next month. ----- > Could you elaborate more about the human > detection algorithm in RFW? How do I access it? Also, maybe you could point me in a direction RFW is a way of connecting your AIBO to a PC through WiFi. AIBO will send notification messages to a PC program, and the PC program can control AIBO remotely. No custom code runs on AIBO itself (a LAN connection is essential) There are semantic input notification SEMID_FACE and SEMID_FACE_ID that tell the PC program when AIBO detects a face. --- The face detection algorithm itself is not adjustable. It is a trade-off. If you want your own algorithm, then download the image to the PC and analyze the photo yourself. The built-in algorithm If you want good integration (I assume the faces detected will also be the ones that trigger key AiboMind2 behavior) --- Try it yourself: Train your AiboMind (V1) memory stick to remember your face. Then use the ABrowser7. Pick "View" and "Human Faces" from the menu. Ignore the "d?" images (I think they are the Japanese ERC developers) Look at the last image, it AIBO's idea of what your face looks like http://www.aibohack.com/ers7/browser.htm |
| V 1420 Api Libaray on OPEN-R bpaul [2004/10/05-22:20] |
| | Response : HomePage : Edit : Delete |
|
| I
intend to create a API on top of OPEN-R using C++ (or Java). | Any suggestions on how to start on this are welcome. I would be a great help ! | |
| +- 1421 RE:Api Libaray on OPEN-R AiboPet [2004/10/06-00:10] |
| | Response : Edit : Delete |
|
| >I
intend to create a API on top of OPEN-R using C++ (or Java). | A C++ API on top of a C++ API... | | >Any suggestions on how to start on this are welcome. | | Start out with the basic question: | "what problem are you trying to solve?" | | And related "who is the audience for your creation?" | | ----- | Example answers: to make it easier to program for newbies, to provide more pre- | built parts, to hide the complexity of the raw OpenR API, to make it easier to | glue together parts, to make it easier to write RCODE like personalities, to | make it easier to write RoboCup programs, to make it easier for robot | experimenters, to fill in missing features (regular walking, voice | recognition, pattern recognition etc), to provide a new and exciting way of | doing X, ... | | NOTE: the answer should not be "all of the above" ;-> | |
| +- 1424 RE:Api Libaray on OPEN-R bpaul [2004/10/06-16:17] |
| | Response : Edit : Delete |
|
| And
related "who is the audience for your creation?" | answer: To fill in missing features (regular walking etc.) related to motion only. | |
| +- 1426 RE:Api Libaray on OPEN-R AiboPet [2004/10/07-01:22] |
| | Response : Edit : Delete |
|
| >And
related "who is the audience for your creation?" | >answer: To fill in missing features (regular walking etc.) related to motion | only. | | Then my suggestion would be to create them as standard OpenR component with | well defined and documented interfaces. They can then be used by anyone (using | the OpenR SDK) | | Good examples: (with some caveats) | All the system builtin objects | The OMWares library (however no source and poor documentation) | The CMU motion module (source, but no standard walking, so-so documentation) | | Object communication is official OpenR object model. I call this using the | OpenR model the way it is intended (not a different "API" on top of OpenR) | | --- | Bad examples: all-or-nothing layers like Tekkotsu - which are APIs on top of | OpenR | (to reuse code you have to 'buy in' to the extra layer -- components written | for Tekkotsu can not be used by regular OpenR programs) | |
| |- 1427 RE:Api Libaray on OPEN-R ejt [2004/10/07-06:37] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Bad
examples: all-or-nothing layers like Tekkotsu - which are APIs on top of | >OpenR | >(to reuse code you have to 'buy in' to the extra layer -- components written | >for Tekkotsu can not be used by regular Open-R programs) | | Other Open-R programs are welcome to use Tekkotsu too (it *is* open source). | You would just have to modify the MMCombo object to add the inter-process | communication you want to provide, which is the same thing you'd be doing anyway | if you were writing an Open-R object from scratch. | | Really, I'd say the best route is to write a common engine that doesn't depend | on any architecture at all, and then that code can be wrapped by either an | Open-R object or Tekkotsu MotionCommand (or other framework) as desired. *shrug* | | If reusability is the main concern, I should also mention that we do intend to | begin porting the system interfaces of Tekkotsu to linux, so that it can compile | for either Aperios (AIBO) or Linux (desktop simulation, or hobbyist robots). | I'll leave it open for debate if anyone can (legally?) use an Open-R | compatability layer on third-party robots. So it depends what kind of | reusability you have in mind. But of course, this cross-platform capability is | still under development and won't be ready for a few more months, so take that | as you will. | |
| +- 1432 RE:Api Libaray on OPEN-R bpaul [2004/10/11-19:57] |
| | Response : Edit : Delete |
|
| Yea,
think i'll try an OpenR component with the object model same as OPEN-R. | >All the system builtin objects | >The OMWares library (however no source and poor documentation) | : CAN'T FIND IT ANYWHERE | >The CMU motion module (source, but no standard walking, so-so documentation) | : Uses the Tekkotsu framework, so NO. | | >I call this using the OpenR model the way it is intended (not a different "API" | on top of OpenR) : Although it would be an attractive option, if a simple model | can be defined. | >--- | As far as Tekkotsu, it has its own set of rules (thats why its a "Framework") | So I dont wanna go into that. I will rather stick to OPEN-R | |
| |- 1435 RE:Api Libaray on OPEN-R ejt [2004/10/12-12:46] |
| | Response : HomePage : Email : Edit : Delete |
|
| >>The
CMU motion module (source, but no standard walking, so-so documentation) | >: Uses the Tekkotsu framework, so NO. | Actually, other way around - Tekkotsu uses the Walk component from CMU motion module, which uses | "normal" open-r communication, so that might fit in with what you want to do. *shrug* | |
| +- 1437 Walking modules (was Api Libaray on OPEN-R) AiboPet [2004/10/13-00:54] |
| Response : Edit : Delete |
|
>>The
OMWares library (however no source and poor documentation) >: CAN'T FIND IT ANYWHERE It is only available for the ERS-210 and 220. Look for the "SoccerLion" sample (in downloads). The "OMWares" set of modules do many things. The most interesting is the "LG" module (leg walking, in styles similar to AiboLife1) BTW: I've asked repeatedly for Sony to minimally port it to the ERS-7, but I'm not holding my breath... >>The CMU motion module (source, but no standard walking, so-so documentation) >: Uses the Tekkotsu framework, so NO. As mentioned by 'ejt' the CMU motion module is a straight OpenR module. It is available in at least 3 forms. The original (CMU Pack 02 release with all their robocup code, ERS-210), the Tekkotsu wrapped version (ERS-2x0 and ERS-7), and my own trimmed down version (ERS-2x0 and ERS-7, trimmed down to core walking) For my own experiments: (all straight Open-R) http://www.aibohack.com/openr_sdk/index.html Walkbench (ERS-2x0 only) has both the OMWare normal walks and the CMU knee walk algorithm. You can switch between them easily. Check out the video. AiboSpeedRacer (ERS-2x0 or ERS-7) has only the CMU knee walk + video feedback. |
| V 1431 How to use EndpointType_SMTP samlu [2004/10/11-10:45] |
| Response : Email : Edit : Delete |
|
In
IPv4 Reference manual, it documents that I can specify the EndpointType_SMTP or EndpointType_POP3 for protocol. Is there any detailed documentation about how to use this (EndpointType_SMTP) for me to send an email out from Aibo? If not, where I can find the sample code about how to send or receive emails via Aibo. |
| V 1389 minor documentation discrepancy: 220 head tilt joint limit ejt [2004/09/18-03:27] |
| | Response : HomePage : Email : Edit : Delete |
|
| Just in case anyone cares... | In section 1.2.1 of the ModelInformation_220_E.pdf, a diagram is shown with the mechanical limits of | the head joints. Note that the minimum angle for the tilt joint is -91.5. However, in the table in | section 2.2.1, the mechanical limit is still listed as -85. I would guess this is a carry over from the 210 | model, which had a limit of -85. | Since software limits have a 3 degree buffer from the mechanical limits, I assume this means the | software limit should also actually be 88.5. | | A whole extra 6.5 degrees to play with! Yay! I'm sure this will just make everyone's day. ;) | |
| |- 1401 RE:minor documentation discrepancy: 220 head tilt joint limit bpaul [2004/09/23-16:40] |
| | | Response : Edit : Delete |
|
| | I
checked out the motion editor. The head tilt T1 is of this range (0,
-75) | | and the head tilt T2 is like this range (45, -15) | | Does that fit ? | | |
| | +- 1402 RE:minor documentation discrepancy: 220 head tilt joint limit ejt [2004/09/23-23:49] |
| | Response : HomePage : Email : Edit : Delete |
|
| >I
checked out the motion editor. The head tilt T1 is of this range (0,
-75) | >and the head tilt T2 is like this range (45, -15) | >Does that fit ? | Two tilt joints? I think you are thinking of the ERS-7, my post was regarding the ERS-220 ;) | (But those are correct for the ERS-7) | |
| +- 1430 RE:minor documentation discrepancy: 220 head tilt joint limi openrsupport@(Administrator) [2004/10/08-04:03] |
| Response : Edit : Delete |
|
Thank
you for reporting the error. The ERS-220 head tilt joint has a
mechanical limit of -91.5 and a software limit of -88. The model information document will be updated in the next release. AIBO SDE Support |
| V 1429 AIBO Remote Framework and AIBO MIND 2 openrsupport@(Administrator) [2004/10/07-09:41] |
| Response : Edit : Delete |
|
ERS-7M2 has been announced. http://news.sel.sony.com/pressrelease/5228 ERS-7M2 includes AIBO MIND 2. AIBO Remote Framework(Final) is the same as AIBO MIND 2! The PC application "AIBO Entertainment Player" (included with ERS-7M2) is written using AIBO Remote Framework. The PC applications made with AIBO Remote Framework can control an AIBO running AIBO MIND 2. The "Memory Stick" of the beta is different from the final version. Please see the FAQs regarding AIBO Remote Framework(Final). https://openr.aibo.com/openr/eng/perm/rfw/regi_faq4.php4#05 Thank you, AIBO SDE Support (The article that 2004/10/07-09:40 posted, was modified by Author) |
| V 1377 Author DELETE ***** [2004/10/05-18:52] |
| | Response : Edit : Delete |
|
| (The
article that 2004/09/15-19:55 posted, was modified by Author) | |
| +- 1418 Author DELETE ***** [2004/10/07-01:16] |
| Response : Edit : Delete |
|
(The
article that 2004/10/05-02:17 posted, was modified by Author) |
| V 1400 what kind of error is Break point exception? aiboKang [2004/09/23-08:03] |
| | Response : Edit : Delete |
|
| Hi, | I've had 9 Break point exception when debuging program for robocupgame. Here | is the emon.log file: | | ------------------------------------------------------------------------------ | mCOOP exception | [exception info] | time stamp: 0x000000081cc44b97 | status:$12: 0x2000ff03, cause:$13: 0x00000024 | badvaddr: $8: 0xe1937cbc, epc:$14: 0x8106b880 | [object info] | context: 0x8023d940, state: 0x00000001 | last context: 0x00000000,last thread: 0x00000000 | [initial value info] | sr:$12: 0x0000ff03 | [register dump] | at: r1: 0x810e28e0, v0: r2: 0x810f64c8, v1: r3: 0x00000000, a0: r4: 0x00000000 | a1: r5: 0x00000000, a2: r6: 0xbc6a7efa, a3: r7: 0x3fd89374, t0: r8: 0x810f64c8 | t1: r9: 0x00000000, t2:r10: 0x0000ff03, t3:r11: 0x00000005, t4:r12: 0x00007f02 | t5:r13: 0x00000002, t6:r14: 0x83f42e08, t7:r15: 0x8023d942, s0:r16: 0x810f7450 | s1:r17: 0x810f7450, s2:r18: 0x810f5ec8, s3:r19: 0x8114b670, s4:r20: 0x8114b870 | s5:r21: 0x810f7480, s6:r22: 0x8114b6a0, s7:r23: 0x810f5a40, t8:r24: 0x83f80000 | t9:r25: 0x81066cc4, gp:r28: 0x810fc2c0, sp:r29: 0x8114b448, s8:r30: 0x8114b910 | ra:r31: 0x8106b864, hi: : 0x00000040, lo: : 0x51eb8557 | | exception code: 9 Break point exception ***************************** !! | | [system control co-processor registers] | Index 0x0000000b Random 0x0000001b EntryLo0 0x01c00317 EntryLo1 0x00000001 | Context 0x00120070 PageMask 0x00000000 Wired 0x0000000c ??? ---------- | BadVAddr 0xe1937cbc Count 0xc21e137f EntryHi 0x80000000 Compare 0x1ce6cc2f | Status 0x2000ff00 Cause 0x1000802c EPC 0x800ab4fc PRId 0x00002751 | Config 0x404664bb LLAddr 0x0007d31a ??? ---------- ??? ---------- | XContext 0x84920070 ??? ---------- ??? ---------- ??? ---------- | ??? ---------- ??? ---------- ECC 0x00000000 CacheErr 0xf0800000 | TagLo 0x00000000 TagHi 0x00000000 ErrorEPC 0xbfc00e14 ??? ---------- | [floating-point general registers] | $f00: 0x00000000 $f01: 0x00000000 $f02: 0x00000000 $f03: 0x00000000 | $f04: 0xbc6a7efa $f05: 0x3fd89374 $f06: 0x00000000 $f07: 0x00000000 | $f08: 0x00000000 $f09: 0x00000000 $f10: 0xffffffff $f11: 0x7ff7ffff | $f12: 0x00000000 $f13: 0x00000000 $f14: 0x00000000 $f15: 0x00000000 | $f16: 0x00000000 $f17: 0x40544000 $f18: 0x00000000 $f19: 0x41086a00 | $f20: 0xffffffff $f21: 0x7ff7ffff $f22: 0xd2f1a9fc $f23: 0x3f80624d | $f24: 0xffffffff $f25: 0x7ff7ffff $f26: 0x00001008 $f27: 0x00000000 | $f28: 0x002c0430 $f29: 0x80800000 $f30: 0x00288026 $f31: 0x00308001 | [floating-point control registers] | $0:fp_eir: 0x00002750 | $31:fp_csr: 0x01000000 (bit[17-0]: ----------------RN) | [object list] | Name ContextID ThreadID ExecSpcID OID MetaSpace | ------------------------------------------------------------------------------- | systemCore 0x8022a5e0 0x8022a780 0x8022a740 ---------- mCore | systemCore(1) 0x8022b3e0 0x8022b540 0x8022a740 ---------- mCore | mCOOPReflector 0x8022ebc0 0x802303c0 0x8022fbc0 0x8000000b mCore | uniMailer 0x8022ed20 0x80230420 0x8022fc00 0x8000000c mCore | mCOOPFaultHandler 0x8022ee80 0x80230480 0x8022fc40 0x8000000d mCore | mDriveFaultHandler 0x8022efe0 0x802304e0 0x8022fc80 0x8000000e mCore | registryManager 0x8022f140 0x80230540 0x8022fcc0 0x8000000f mCore | addressManager 0x8022f2a0 0x802305a0 0x8022fd00 0x80000010 mCore | kernelModeLib 0x8022f400 0x80230600 0x8022fd40 0x80000011 mCore | mCoreReflector 0x8022f560 0x80230660 0x8022fd80 0x80000012 mCore | idle 0x8022f6c0 0x802306c0 0x8022fdc0 0x80000013 mCore | exceptionHandler 0x8022f820 0x80230720 0x8022fe00 0x80000014 mCore | analyzer 0x8022f980 0x80230780 0x8022fe40 0x80000015 mCore | mDriveReflector 0x80234980 0x802307e0 0x8022fe80 0x80000016 mCore | obletManager 0x80234820 0x80230840 0x8022fec0 0x80000017 mCore | installer 0x802346c0 0x802308a0 0x8022ff00 0x80000018 mCore | driverManager 0x80234560 0x80230900 0x8022ff40 0x80000019 mCore | objectManager 0x80234400 0x80230960 0x8022ff80 0x8000001a mCore | mClassReflector 0x802342a0 0x802309c0 0x8022ffc0 0x8000001b mCore | mSystemReflector 0x80234140 0x80230a20 0x80230000 0x8000001c mCore | mAVFaultHandler 0x80233fe0 0x80230a80 0x80230040 0x8000001d mCore | mAVReflector 0x80233e80 0x80230ae0 0x80230080 0x8000001e mCore | mAVInit 0x80233d20 0x80230b40 0x802300c0 0x8000001f mCore | avManager 0x80233bc0 0x80235340 0x80230100 0x80000020 mCore | eventManager 0x80238180 0x802352e0 0x80230140 0x80000021 mCore | downloader 0x80237c00 0x80235160 0x80230240 0x80000028 mSystem | nonBlockingFileSystemManager 0x80237aa0 0x80235100 0x80230280 0x80000029 mDrive | openrBusManager 0x80237940 0x802350a0 0x802302c0 0x8000002a mDrive | openrBusManager(1) 0x80238bc0 0x8023c2e0 0x802302c0 0x8000002a mDrive | bmnDriver 0x802377e0 0x80235040 0x80230300 0x8000002b mDrive | bmnDriver(1) 0x802392a0 0x80234ce0 0x80230300 0x8000002b mDrive | fbkDriver 0x80237680 0x80234fe0 0x80230340 0x8000002c mDrive | fbkDriver(1) 0x8023d520 0x80240ae0 0x80230340 0x8000002c mDrive | fbkDriver(2) 0x8023d3c0 0x80240a80 0x80230340 0x8000002c mDrive | cardManager 0x80237520 0x80234f80 0x80230380 0x8000002d mDrive | cardManager(1) 0x8023c680 0x8023bf80 0x80230380 0x8000002d mDrive | memoryStickDriver 0x802373c0 0x80234f20 0x80238b80 0x8000002e mDrive | memoryStickDriver(1) 0x80239400 0x80234d40 0x80238b80 0x8000002e mDrive | fatFileSystem 0x80239980 0x80234ec0 0x80238b40 0x8000002f mCOOP | memoryStickWatcher 0x80239820 0x80234e60 0x80238b00 0x80000030 mDrive | oobjectManager 0x802396c0 0x80234e00 0x80238ac0 0x80000031 mCOOP | aperiosClass 0x80239560 0x80234da0 0x80238a80 0x80000032 mClass | opowerManager 0x80238020 0x80235280 0x80230180 0x80000033 mCOOP | oserviceManager 0x80237ec0 0x80235220 0x802301c0 0x80000034 mCOOP | ovirtualRobot 0x80237d60 0x802351c0 0x80230200 0x80000035 mCOOP | odesignedRobot 0x80239140 0x80234c80 0x80238a40 0x80000036 mCOOP | osystemLogger 0x80238fe0 0x80234c20 0x80238a00 0x80000037 mCOOP | ovirtualRobotComm 0x80238e80 0x80234bc0 0x802389c0 0x80000038 mCOOP | ovirtualRobotAudioComm 0x80238d20 0x8023c340 0x80238980 0x80000039 mCOOP | IPStack 0x8023d180 0x8023c280 0x80238940 0x8000003a mDrive | OrinocoDriver 0x8023d020 0x8023c220 0x80238900 0x8000003b mDrive | OrinocoDriver(1) 0x8023c520 0x8023bf20 0x80238900 0x8000003b mDrive | OrinocoEnabler 0x8023cec0 0x8023c1c0 0x802388c0 0x8000003c mDrive | hookConsoleIO 0x8023cd60 0x8023c160 0x80238880 0x8000003d mDrive | emergencyMonitor 0x8023cc00 0x8023c100 0x80238840 0x8000003e mCOOP | netconf 0x8023caa0 0x8023c0a0 0x80238800 0x8000003f mCOOP | anttcpio 0x8023c940 0x8023c040 0x802387c0 0x80000040 mDrive | hookConsoleIOActivator 0x8023c7e0 0x8023bfe0 0x80238780 0x80000041 mCOOP | powerMonitor 0x8023c3c0 0x8023bec0 0x80238740 0x80000042 mCOOP | tcpgw 0x8023e180 0x8023be60 0x80238700 0x80000043 mCOOP | perception 0x8023e020 0x8023be00 0x802386c0 0x80000044 mCOOP | localization 0x8023dec0 0x8023bda0 0x80238680 0x80000045 mCOOP | control 0x8023dd60 0x8023bd40 0x80238640 0x80000046 mCOOP | behavior 0x8023dc00 0x8023bce0 0x80238600 0x80000047 mCOOP | shMemMgr 0x8023daa0 0x8023bc80 0x802385c0 0x80000048 mCOOP | motion 0x8023d940 0x8023bc20 0x80238580 0x80000049 mCOOP | tinyFTPD 0x8023d7e0 0x8023bbc0 0x80238540 0x8000004a mCOOP | w3aibo 0x8023d680 0x80240b40 0x80238500 0x8000004b mCOOP | [stack info] | area: 0x811239e8 - 0x8114b9e8, sp: 0x8114b448 | [stack dump] | 8114b408: 810fc2c0 81066f34 810fc2c0 4036adcc ....4o........6@ | 8114b418: 76a3fd44 c0569840 810fc2c0 c0515a43 D..v@.V.....CZQ. | 8114b428: 810f7450 810f7450 810fc2c0 8106b820 Pt..Pt...... ... | 8114b438: ffffffff 7ff7ffff d2f1a9fc 3f80624d ............Mb.? | 8114b448: c75cc580 c05e3efb bc6a7efa 3fd89374 ..\..>^..~j.t..? | 8114b458: 00000000 00000000 00000000 00000000 ................ | 8114b468: 0ef870cd c0519859 b9215760 4051cc95 .p..Y.Q.`W!...Q@ | 8114b478: 00000000 00000000 00000000 00000000 ................ | 8114b488: 00000000 00000000 00000000 40590000 ..............Y@ | 8114b498: 00000000 00000000 00000000 00000000 ................ | 8114b4a8: 00000000 c062c000 810fc2c0 c050a01c ......b.......P. | 8114b4b8: 00000000 00000000 00000002 00000008 ................ | 8114b4c8: 00000100 00000038 00000000 810e47c0 ....8........G.. | 8114b4d8: 816f0069 40284504 00000000 00000000 i.o..E(@........ | 8114b4e8: bc6a7efa 3fd89374 00000000 00000000 .~j.t..?........ | 8114b4f8: 00000000 00000000 00000000 00000000 ................ | * | 8114b518: 810fc2c0 8114b910 810d866c 00000000 ........l....... | 8114b528: 810e66d0 8114b570 810fc2c0 8114b648 .f..p.......H... | 8114b538: 810fc2c0 00000000 000000f4 810e0a0c ................ | 8114b548: ffffffff 7ff7ffff ffffffff 7ff7ffff ................ | 8114b558: ffffffff 7ff7ffff bc6a7efa 3fd89374 .........~j.t..? | 8114b568: 00000000 00000000 00000000 00000000 ................ | 8114b578: 000000bb 00000208 810f8dac 000000f3 ................ | 8114b588: 00000000 00000000 810f8da8 80000037 ............7... | 8114b598: 00000000 00000000 00000000 00000000 ................ | 8114b5a8: 810f64c8 80000037 00000004 810f8da0 .d..7........... | 8114b5b8: 00000100 8114b5f8 00000008 80000022 ............"... | 8114b5c8: 8114b560 00000000 81084550 81084674 `.......PE..tF.. | 8114b5d8: 810fc2c0 00000000 810fc2c0 8106b444 ............D... | 8114b5e8: 00000000 810f7450 810f5ec8 8114b670 ....Pt...^..p... | 8114b5f8: 8114b870 810f7480 8114b6a0 810f5a40 p....t......@Z.. | 8114b608: 810fc2c0 8114b910 8106f09c 81084774 ............tG.. | 8114b618: 00000000 408f4000 32b1dc71 3faca78c .....@.@q..2...? | 8114b628: ffffa786 ffffffff 00001008 00000000 ................ | 8114b638: 810f64c8 bfcad5f1 8114b670 40284504 .d......p....E(@ | 8114b648: ded289d4 bff664ee 00000000 00000000 .....d.......... | 8114b658: 00000000 3ff80000 00000000 3ff2ee0b .......?.......? | 8114b668: 810fc2c0 3f722000 00000000 00000000 ..... r?........ | 8114b678: 00000000 00000000 00000000 40574000 .............@W@ | 8114b688: 00000000 00000000 a2529d39 3fc1df46 ........9.R.F..? | 8114b698: 00000000 00000000 00000003 6a7b8c9d ..............{j | 8114b6a8: 08453556 40605d17 6f859024 4050a0e5 V5E..]`@$..o..P@ | 8114b6b8: 00000000 00000000 7e28d220 3f97ec41 ........ .(~A..? | 8114b6c8: f87027b0 3fb6f213 7eb8fc0b 3ffa82f3 .'p....?...~...? | 8114b6d8: 00000003 810d0f80 64feaa46 405a08ed ........F..d..Z@ | 8114b6e8: de5dab34 c050a74f 8f5c28fe 402f33c2 4.].O.P..(\..3/@ | 8114b6f8: fabb0904 bfdc05c6 ec156ae0 3fca392a .........j..*9.? | 8114b708: 7bcaad1d 40023834 00000003 6a7b8c9d ...{48.@......{j | 8114b718: dbcb5bd2 c05d7a18 5dbaa903 40510081 .[...z]....]..Q@ | 8114b728: ccccccd3 4036adcc bf535c26 bfe30ade ......6@&\S..... | 8114b738: 060594f8 3fc786f2 3e97192b 4000650f .......?+..>.e.@ | 8114b748: 00000003 810cf49c 76a3fd44 c0569840 ........D..v@.V. | 8114b758: ab92b10e c0515a43 00000000 00000000 ....CZQ......... | 8114b768: f5d22cce bfeb00bf 2bf40d98 3fc5a638 .,.........+8..? | 8114b778: e32df58e 3ffe3653 00000001 00000000 ..-.S6.?........ | 8114b788: 00000000 00000000 00000000 00000000 ................ | 8114b798: 00000000 00000000 1cfb2ee0 bfd618a8 ................ | 8114b7a8: 38bcdff0 bfcad5f1 00000000 00000000 ...8............ | 8114b7b8: 00000001 00000000 00000000 00000000 ................ | 8114b7c8: 00000000 00000000 00000000 00000000 ................ | * | 8114b7f8: 810fc2c0 8107e730 810fc2c0 00000002 ....0........... | 8114b808: 810fc2c0 00000000 810fc2c0 81077308 .............s.. | 8114b818: 00000001 000197c8 810f5b14 00000002 .........[...... | 8114b828: 8114b870 806d7864 00000000 810fc2c0 p...dxm......... | 8114b838: 810727f0 810773a4 00000000 408f4000 .'...s.......@.@ | 8114b848: 32b1dc71 3faca78c ffffa786 ffffffff q..2...?........ | 8114b858: 00000018 00000000 8107b488 8107b500 ................ | 8114b868: 810fc2c0 8108004c 7e28d220 3f97ec41 ....L... .(~A..? | 8114b878: f87027b0 3fb6f213 7eb8fc0b 3ffa82f3 .'p....?...~...? | 8114b888: fabb0904 bfdc05c6 ec156ae0 3fca392a .........j..*9.? | 8114b898: 7bcaad1d 40023834 bf535c26 bfe30ade ...{48.@&\S..... | 8114b8a8: 060594f8 3fc786f2 3e97192b 4000650f .......?+..>.e.@ | 8114b8b8: f5d22cce bfeb00bf 2bf40d98 3fc5a638 .,.........+8..? | 8114b8c8: e32df58e 3ffe3653 1cfb2ee0 bfd618a8 ..-.S6.?........ | 8114b8d8: 38bcdff0 bfcad5f1 00000000 00000000 ...8............ | 8114b8e8: 00000000 00000000 00000000 00000000 ................ | 8114b8f8: 00000000 00000000 806d8190 8114b9d0 ..........m..... | 8114b908: 8114b9d8 810fc2c0 81070101 8107d6c0 ................ | 8114b918: 00000000 00030000 057619f1 00000000 ..........v..... | 8114b928: 810f5b14 00000001 8063d068 805dfa40 .[......h.c.@.]. | 8114b938: 696e6974 80100860 805df9e0 00000000 tini`.....]..... | 8114b948: 810fc2c0 00000000 810721b8 41e00000 .........!.....A | 8114b958: 00000001 80000038 00000005 00000004 ....8........... | 8114b968: 810fc2c0 00000001 8114b9d0 810fc2c0 ................ | 8114b978: 81063034 81062fe0 00000004 00000002 40.../.......... | 8114b988: 80000031 802a0198 810fc2c0 802a022c 1.....*.....,.*. | 8114b998: 00000000 80000038 00000007 00000003 ....8........... | 8114b9a8: 80000038 00000007 8114b958 810fc2c0 8.......X....... | 8114b9b8: 810637d8 80226688 0000000a 802a0228 .7...f".....(.*. | 8114b9c8: 00000106 80000038 00000105 00000003 ....8........... | 8114b9d8: 80000038 00000007 00000001 806d7864 8...........dxm. | | According to the above emon.log file, cpu exception of Break point happened. | Being short of knowledge about MIPS, I'm not sure what kind of error the Break | point exception is , which is not mentioned in ProgrammerGuide_E.pdf document, | and how to fix it. | | Could anybody give me any ideas? Thanks a lot in advance. | |
| |- 1416 RE:what kind of error is Break point exception? openrsupport@(Administrator) [2004/10/05-02:01] |
| | Response : Edit : Delete |
|
| According
to your emon.log, the exception is occurring in the motion object. | Because this is a robocup object, hopefully someone involved with robocup will | be able to help you. | | For more details on CPU exceptions, please refer to the MIPS architecture | documents. (You can find them by searching the WEB for "MIPS Architecture | Documentation".) | | AIBO SDE Support | |
| +- 1422 RE:what kind of error is Break point exception? Grigg [2004/10/06-02:20] |
| Response : Edit : Delete |
|
It
seems an integer-division-by-zero has occurred. The compiler may emit a div-by-zero checking sequence with a breakpoint instruction. |
| V 1419 Poll question bpaul [2004/10/05-17:23] |
| Response : HomePage : Edit : Delete |
|
Please
feel free to take part in this poll or join the group itself. It will
ask your preference when it comes to AIBO programming. http://groups.yahoo.com/group/enstaopenrtutorial/surveys?id=11890209 |
| V 1408 How to prevent Aibo's automatic shutdown ? aiboKang [2004/09/30-00:40] |
| | Response : Edit : Delete |
|
| Hi,everyone | My ERS-7 for Robocup always has the problem of automatic shutdown. | In Technical FAQ of this website,it says that "You can monitor the value of | the battery current with OPENR::GetPowerStatus()" to prevent AIBO's shutdown | due to 'BATTERY_OVER_CURRENT' , but I still want to know how to solve it | technically. | How does it happen? Is that because of 'BATTERY_OVER_CURRENT' or anything | else? | Can anyone help me with the detailed solution? It's needed urgently!! | Thanks very much! | (The article that 2004/09/30-00:40 posted, was modified by Author) | |
| +- 1417 RE:How to prevent Aibo's automatic shutdown ? openrsupport@(Administrator) [2004/10/05-02:03] |
| Response : Edit : Delete |
|
AIBO
will shutdown if too much current is being drawn from the battery. This may happen if the motors are being driven very hard. The shutdown is a safety mechanism to prevent damage to AIBO. By monitoring the battery current with OPENR::GetPowerStatus(), you can prevent the BATTERY_OVER_CURRENT shutdown by reducing power consumption when the battery current draw is high. This can be done by shutting off power to the motors, or turning down the PID values for AIBO's joints. Thank you, AIBO SDE Support |
| V 1382 problem with ObjectComm-multi rosaluna [2004/09/15-20:02] |
| | Response : Edit : Delete |
|
| I have a problem with ObjectComm-multi. | I have compiled and executed sucessfully ObjectComm-multi/RP3. | I have compiled sucessfully ObjectComm-multi/RP2 but I have problem with the | execution: | after that the connection is established with all gates, my aibo "dies"!. | Why? | I must change something in the WLANCONF.TXT or in the WLANDFLT.TXT? | |
| +- 1415 RE:problem with ObjectComm-multi openrsupport@(Administrator) [2004/10/05-01:52] |
| Response : Edit : Delete |
|
WLANDFLT.TXT
contains default wireless connection settings. You will only need to edit WLANCONF.TXT. If you were successful in running ObjectComm-multi/RP2, then you shouldn't have a problem with RP3. Using the same Memory Stick as RP2, you should be able to compile and install RP3/robot, and copy RP3/robot/MS to your Memory Stick. Then boot and telnet into AIBO. Compile RP3/host (make sure HOSTGW.CFG is correct), run ipc-daemon, and then run start-rp-openr. You can look at the installation guide to make sure you are running Remote Processing OPEN-R correctly. If you continue to have problems, please post with more detail about what is going wrong. Thank you, AIBO SDE Support |
| V 1412 Remote Framework --- Track, Look & Search Fairlawn07 [2004/10/01-11:34] |
| | Response : Email : Edit : Delete |
|
| I have questions about the following in RemoteTest sample: | 1. I have problem to understand: Track HV, Track Face HV & Look Relativity HV. | 2. What is the difference between Look Target and Search Target. | |
| +- 1414 RE:Remote Framework --- Track, Look & Search openrsupport@(Administrator) [2004/10/02-02:22] |
| Response : Edit : Delete |
|
>1.
I have problem to understand: Track HV, Track Face HV & Look
Relativity HV. >2. What is the difference between Look Target and Search Target. HV means Horizontal and Vertical. Track HV -- Move head to the specified position (HV degrees) and track the target (keep the target centered in the camera image). Track Face HV -- Move head to the specified position (HV degrees) and track a face (keep the face centered in the camera image). Look Relativity HV -- Move head to the specified position (HV relative degrees) and look at the target (move the camera once so the target is centered). HV relative degrees means if AIBO's head HV is (30, 0) and you try to set the relative degrees (20, 0), AIBO's final head HV will be (50, 0). Look Target -- If the target is in view, move the head so the target is centered in the camera image. AIBO will not move its head if it cannot see the target. Search Target -- Move the head around until AIBO can see the target. If AIBO finds the target, the head will move to center the target in the camera image. AIBO SDE Support |
| V 1410 AIBO Network Authentication nelsonc [2004/09/30-13:59] |
| | Response : Email : Edit : Delete |
|
| Is
there are way to set-up the ABO network via WPA-TSK network
authentication | and TKIP data encryption? Thanks all! | |
| +- 1413 RE:AIBO Network Authentication openrsupport@(Administrator) [2004/10/02-01:19] |
| Response : Edit : Delete |
|
I'm
sorry, but there is no way to set AIBO's authorization mode to "WPA", "WPA-PSK", or "Shared Key" authentication. If possible, please set the authorization mode to "Open System Authorization". You can specify either WEP-encrypted or unencrypted network communication. AIBO SDE Support |
| V 1403 R-Code arithmetic error found? outsider [2004/09/24-18:09] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi, | I think I have found an error inside R-Code related to its arithmetic (only | tested with an ERS-7 model).Let me introduce the following code: | | EDIT | LET:number:507354182 | IF:number:>:0:THEN | PRINT:"Number greater than zero" | ELSE | PRINT:"Number under zero" | ENDIF | END | | Looking at the code above, you can see that the message that must be posted on | the telnet console must be "Number greater than zero", but instead, if you | execute this code on your ERS-7, the message posted will be "Number under zero". | Is it there any explanation appart from being a failure in R-Code design?. | It is suppossed that numbers in ERS-7 are 32 bits long (31 bits for the number | and 1 bit for the sign), so the maximum number is 2147483647 (i.e. 2^31 - 1). We | must be able to use numbers from -2147483647 to +2147483647. I don't understand | why R-Code detects the number 507354182 as being below zero, but 507254182 as | being greater than zero. | Where is the error?. Is it my understanding error or is it an R-Code error?. | | Many thanks | |
| +- 1404 RE:R-Code arithmetic error found? AiboPet [2004/09/25-00:48] |
| | Response : Edit : Delete |
|
| It
does the same thing with "32768" (treated as negative) | | An ERS-7 specific RCODE interpreter bug | (not a design error, blame the old 16 bit implementation) | | They appear to be testing the wrong bit (bit 15) instead of the proper sign | bit (bit 31) | ---- | RCODE for other models (11x, 2x0, 31x) is limited to 16 bits (-32768 -> 32767) | With the ERS-7 they bumped the variables to 32 bits, and obviously missed this | case. | | Advice: keep your RCODE variables to sign + 15 bits (-32768 -> 32767) | Of course Sony should fix the bug. | |
| +- 1406 RE:R-Code arithmetic error found? outsider [2004/09/27-23:30] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Advice:
keep your RCODE variables to sign + 15 bits (-32768 -> 32767) | >Of course Sony should fix the bug. | | Thanks Aibopet for your confirmation and advice, but I was implementing floating | poit arithmetics using the IEE 754 standar, which is 32 bits long. I was almos | finished the addition of two numbers and planning to implement the | multiplication and exponentiation. My idea was to release that routines to the | public (as has been long time requested by many R-Code programmers), but now I | will have to look for another solution since I cannot wait for a new version of | R-Code with that fixed. | | Still waiting official confirmation from Sony. Please, Sony's people, can you | confirm that issue and tell us if you are going to release an update for R-Code | on ERS-7 fixing that bug? | | Thanks a lot | |
| +- 1409 RE:R-Code arithmetic error found? openrsupport@(Administrator) [2004/09/30-09:25] |
| | Response : Edit : Delete |
|
| Thank
you very much for reporting the bug. We will try to fix this problem. | However, we do not have a release date for the next R-CODE update. We will | inform you on this website when it is available. Thank you. | | AIBO SDE Support | |
| +- 1411 Thanks outsider [2004/09/30-19:23] |
| Response : HomePage : Email : Edit : Delete |
|
>Thank
you very much for reporting the bug. We will try to fix this problem. >However, we do not have a release date for the next R-CODE update. We will >inform you on this website when it is available. Thank you. > >AIBO SDE Support OK, thank you very much. I will be waiting for the update. |
| V 1405 custom motions in remote framework zpavlov [2004/09/27-14:25] |
| | Response : Email : Edit : Delete |
|
| Hi,
I'm having a problem playing custom motions using remote framework. This | is mainly because the ers-7.mwc file is completely different from the r-code | version. Is there a way I can modify the RFW middle ware command file such | that I can use my own motions? | | Thanks, | Zac | |
| +- 1407 RE:custom motions in remote framework zpavlov [2004/09/28-05:38] |
| Response : Edit : Delete |
|
nvm, I figured it out. >Hi, I'm having a problem playing custom motions using remote framework. This >is mainly because the ers-7.mwc file is completely different from the r-code >version. Is there a way I can modify the RFW middle ware command file such >that I can use my own motions? > >Thanks, >Zac |
| V 1391 Java API bpaul [2004/09/20-21:59] |
| | Response : Edit : Delete |
|
| I
would like to program Aibo using java, but i can't seem to find an API
build | over the OPEN-R sdk, that could help me with my goal. | | Does anyone kow any remote framework/ interface to Aibo that can be programmed | in java. | |
| +- 1393 RE:Java API AiboPet [2004/09/21-02:09] |
| | Response : HomePage : Edit : Delete |
|
| >I
would like to program Aibo using java, but i can't seem to find an API
build | >over the OPEN-R sdk, that could help me with my goal. | > | >Does anyone kow any remote framework/ interface to Aibo that can be programmed | >in java. | | There is no solution for running Java code on the AIBO itself (ie. no Java | interpreter or runtime on the dog itself) | | There are several solutions for running a PC based Java program and connecting | it to AIBO using WiFi | It depends on the model of AIBO and what software you run on the dog itself | (ie. not Java) | | There are several samples that use RCode/RCodePlus (dog runs RCODE/RCodePlus) | Tekkotsu has several such programs (dog runs OpenR SDK C++ code using Tekkotsu) | | There is no Java interface to the Remote Framework (ERS-7 only) but a DLL | wrapper could be written. | | Several questions: what model AIBO ? Is running Java on a PC acceptable ? | What services do you need running on the dog itself ? (eg: standard walking, | voice recognition, audio playback ?) | |
| +- 1394 RE:Java API bpaul [2004/09/21-03:56] |
| | Response : Edit : Delete |
|
| >There
is no solution for running Java code on the AIBO itself (ie. no Java | >interpreter or runtime on the dog itself) | | That's really bad news. | | >There are several solutions for running a PC based Java program and connecting | >it to AIBO using WiFi. It depends on the model of AIBO and what software you | run on the dog itself | >(ie. not Java) | | MODEL: ERS-7 | | >There are several samples that use RCode/RCodePlus (dog runs RCODE/RCodePlus) | >Tekkotsu has several such programs (dog runs OpenR SDK C++ code using | Tekkotsu) | | I don't wanna use R-code or try exploring Tekkotsy (as the whole thing is in | C++, the language i'm trying to avoid) | | >There is no Java interface to the Remote Framework (ERS-7 only) but a DLL | >wrapper could be written. | | I would'nt know how to do that, but i can try exploring. | | >Several questions: | | >what model AIBO ? ERS-7 | | >Is running Java on a PC acceptable ? Well if there is no other possibilty. | | >What services do you need running on the dog itself ? (eg: standard walking, | >voice recognition, audio playback ?) | | I would like to do some advanced behaviorial coding involving everything you | mentioned and more.. (its only possible in Open-R as far as i know but i would | really prefer java) | | IS THERE A WAY OUT FOR ME ? Some accompnying links would be helpful. Perhaps | little more elaboration "PC based Java program and connecting it to AIBO using | WiFi" would be helpful. Really appreciate the reply. | | Regards/ Paul | |
| +- 1395 URBI sapient [2004/09/21-20:56] |
| | Response : Edit : Delete |
|
| >IS
THERE A WAY OUT FOR ME ? | | You may be dismayed to hear this but you'll probably have to get your hands | dirty at some point... I would advise using the ARF(Aibo Remote Framework) and | writing a dll wrapper this is good because it allows you access to lots of the | Sony proprietry libraries that have functions from the commercial software (such | as recharging) that have been a fraustration with open-r | | If this sounds too ominous there is a project here in Paris by Jean-Christophe | Baille, it is basically the same idea as the ARF but written in open-r code, and | runs exclusively on Unix and derivatives (maybe in cygwin?). The link to source | forge is : | | http://sourceforge.net/projects/urbi/ | | The program (on the memstick) is in C/C++ but that wouldn't matter. you could | just look at the "URBI" protocol then send the low-level commands to the aibo by | telnet port 54000 using the very nice Java network commands. The latency is | pretty good, (about as good as the ARF) but it is at a very low level (joint | level control) so you'd have to do a bit of work. (that said, I managed to get | a reasonable walk running across a good WiFi network in about a day) | | I don't know if the project will continue too much now that the ARF is up and | running though. My guess is that you would be better off with ARF. | | Sorry. | |
| +- 1396 RE:URBI bpaul [2004/09/21-23:23] |
| | Response : Edit : Delete |
|
| Think
u have put me on my place. | I'll try that, wonder about alternatives on Mac OSX ? | | So java will be confined to the PC but, never mind. | Thanks anyway | |
| +- 1397 RE:URBI AiboPet [2004/09/22-00:42] |
| | Response : HomePage : Edit : Delete |
|
| Also,
the Remote Framework has the coolest set of features (voice recognition, | walking *AND* pattern recognition) | | > wonder about alternatives on Mac OSX ? | Unfortunately the Remote Framework relies on PC based DLLs. That may work on | the Mac under a Windows emulator (I don't know if anyone has tried it) | | ======= | I know a lot of people don't want to use RCODE, but I *strongly* suggest | looking at it first before anything else. It gives you a lot of canned | behavior with zero work. | Try YART7 http://www.aibohack.com/ers7/yart7.htm | | ========= | Using telnet connected RCODE (this is what I recommend) | | If you are happy with the features of RCODE, but want to program the logic in | Java running on your Mac, then it is easy: | | Write any code you want in Java on the Mac, and when you want AIBO to do | something, you send a simple RCODE string to the dog, eg: | "PLAY ACTION WALK 0 100" // AIBO will walk .1 metre forward | "PLAY ACTION L_CONTACT.M_RIGHT.S_MRKICK" // play canned skit - kicking | | To get data from AIBO, you send a string like | "PRINT "%d %d %d" BattRest Pink_Ball Distance" | Then parse the reply string. | | All trivial LAN socket string code, extremely portable. | | See the following for working Java code that does exactly this. The sample is | somewhat old, for the 2x0 (by 'hkora11' a Japanese developer) | http://www.yk.rim.or.jp/~hkora11/aibo_2nd-g/r-code/controler.htm | |
| +- 1398 RCODE via LAN AiboPet [2004/09/22-00:45] |
| | Response : HomePage : Edit : Delete |
|
| Very
simple example: | | http://www.yk.rim.or.jp/~hkora11/aibo_2nd-g/r-code/commander.htm | |
| +- 1399 RE:RCODE via LAN bpaul [2004/09/22-17:03] |
| Response : Edit : Delete |
|
I
have never gone through R-Code, I will do that now. At the moment i think its "embeded Rcode in java" Correct me if i'm wrong. |
| V 1385 ImageObserver bpaul [2004/09/16-19:00] |
| | Response : Edit : Delete |
|
| Hi all, | | I am currently busy with tweaking the "ImageObserver" sample program that comes | with OPEN-R. Its stub file says: | ---------- | ObjectName : ImageObserver | NumOfOSubject : 1 | NumOfOObserver : 1 | Service : "ImageObserver.Image.OFbkImageVectorData.O", null, Notify() | Service : "ImageObserver.DummySubject.DoNotConnect.S", null, null | ----------------- | I donot understand why there is a DummySubject when "OVirtualRobotComm" (system | object) is the subject in this case | | Thanks beforehand | Paul | |
| +- 1386 RE:ImageObserver AiboPet [2004/09/16-23:48] |
| | Response : HomePage : Edit : Delete |
|
| Two answers: | >I donot understand why there is a DummySubject when "OVirtualRobotComm" | (system | >object) is the subject in this case | | The generation tools and macros don't handle the case of ZERO observers or | ZERO subjects. | So if you have ZERO subjects, the dummy line acts like a placeholder (notice | the last function name argument is 'null') | When adding your first real subject, remove the Dummy line and add a regular | subject, eg: | Service : "ImageObserver.MySubject.MyThang.S", null, MyReadyThang() | When adding the second subject, bump the 'NumOfOSubject' count to 2, and add | another line. | | ---- | NOTE: those are the subjects and observers for that one object (ImageObserver) | There is no subject in this case. | | As you point out, the CONNECT.CFG connects the observer of your ImageObserver | object to the subject of the system's OVirtualRobotComm - however the | ImageObserver object doesn't know about that. | | the "observer" of one object connects to the "subject" of a *different* object | and vice-versa - the "subject" of one object connects to the "observer" of a | different object | |
| +- 1390 RE:ImageObserver bpaul [2004/09/20-17:16] |
| | Response : Edit : Delete |
|
| >NOTE:
those are the subjects and observers for that one object (ImageObserver) | >There is no subject in this case. | > | >As you point out, the CONNECT.CFG connects the observer of your ImageObserver | >object to the subject of the system's OVirtualRobotComm - however the | >ImageObserver object doesn't know about that. | | [CONCLUSION: Does that mean that every system object(that acts as a subject) | ought to be ignored, as the obj-sub connection goes automatically ?] | | >the "observer" of one object connects to the "subject" of a *different* object | >and vice-versa - the "subject" of one object connects to the "observer" of a | >different object | | IN SHORT: So the absense of the Ready() method indicates an absence of a | subject. | |
| +- 1392 RE:ImageObserver AiboPet [2004/09/21-02:02] |
| Response : HomePage : Edit : Delete |
|
>[CONCLUSION:
Does that mean that every system object(that acts as a subject) >ought to be ignored, as the obj-sub connection goes automatically ?] I don't fully understand what you are saying. For system objects you want to use (as specified in CONNECT.CFG) your side of the puzzle is writing the observer code for your side of the connection. There are other cases where you must write the subject (not in ImageObserver) To see which subjects and observers are actually used, read the CONNECT.CFG file. >IN SHORT: So the absense of the Ready() method indicates an absence of a >subject. No, not always. In the dummy case, yes (ie. a subject is there, it is not connected) To see which subjects and observers are actually used, read the CONNECT.CFG file. Having a Ready() method is optional. This callback is called when *your* subject becomes ready. It isn't always needed. ====== The subject/observer connections are used in several different ways. Think of it as plumbing pipes - that can be hooked up in several different ways. These general conclusions may or may not apply to your case. I find it best to discuss the use of subject/observer in a very specific context. In the ImageObserver case (one of the more simple uses), the observer gets image data from system. No subject needed on your side (just the dummy case), no accessing of the subject[] array, no SetData(), etc. If you have a different case or problem, please include the [modified] stub.cfg file and parts of CONNECT.CFG - those decide how the plumbing is wired up for the specific case. |
| V 1374 Convert Remote Framework to .NET component susanto [2004/09/11-22:42] |
| | Response : Edit : Delete |
|
| Hi, | Is it possible to wrap the Remote Framework become .NET component, the reason | is because I using .NET. Can somebody may be build the component? I find it | will be more exciting, since people can used any different programming | language. Thanks. | |
| +- 1376 RE:Convert Remote Framework to .NET component AiboPet [2004/09/13-23:53] |
| | Response : HomePage : Edit : Delete |
|
| >Is
it possible to wrap the Remote Framework become .NET component | | Of course. It could be done in several different ways. It has been requested | before (under the context of Visual Basic support). I don't think Sony is | planning on official support. | The "CVAIBO" class is the main important one (the others are optional) | | BTW: Depending on the tool, you can use imported DLLs as-is, or as Windows | messages. | | >I find it | >will be more exciting, since people can used any different programming | >language. Thanks. | | What language are you using ? / planning on using ? | [For Visual Basic programming, personally I prefer simpler VB DLL wrappers] | | ======= | BTW: the best way to learn about the RFW API is to look at the samples | (written using Visual C++). | |
| +- 1380 RE:Convert Remote Framework to .NET component susanto [2004/09/14-14:23] |
| | Response : Edit : Delete |
|
| Hi
AiboPet, Can you help me to wrap AIBO Remote Framework to .NET
component? So | I just need to 'reference' to my VB.NET application using Visual Studio.NET | 2003 IDE. | I don't want to waste my time to learn C++, which I find very difficult to | understand. | | Thanks. | | I am using Visual Basic.NET language. | >>Is it possible to wrap the Remote Framework become .NET component | > | >Of course. It could be done in several different ways. It has been requested | >before (under the context of Visual Basic support). I don't think Sony is | >planning on official support. | >The "CVAIBO" class is the main important one (the others are optional) | > | >BTW: Depending on the tool, you can use imported DLLs as-is, or as Windows | >messages. | > | >>I find it | >>will be more exciting, since people can used any different programming | >>language. Thanks. | > | >What language are you using ? / planning on using ? | >[For Visual Basic programming, personally I prefer simpler VB DLL wrappers] | > | >======= | >BTW: the best way to learn about the RFW API is to look at the samples | >(written using Visual C++). | |
| +- 1388 RE:Convert Remote Framework to .NET component AiboPet [2004/09/17-00:00] |
| Response : HomePage : Edit : Delete |
|
>Hi
AiboPet, Can you help me to wrap AIBO Remote Framework to .NET
component? It is on my list, but not very high priority [I did a VB wrapper for older stuff before, but there was very little interest] I suggest giving it a go yourself, if you have some problems send me an email [aibopet@aibohack.com]. Just worry about the CVAIBO class. MSDN gives you tips for wrapping C++ classes for C#/NET. The callbacks/notifications will be the tricky part. NOTE: requires a C++ compiler to compile up the wrapper object - the free C++ compiler should be fine Probably more work in creating a useful VB NET sample than in the wrapping the interface itself. The ultimate sample app (RemoteTest in VC++) is pretty big. |
| V 1372 Author DELETE ***** [2004/09/16-03:13] |
| Response : Edit : Delete |
|
(The
article that 2004/09/10-12:24 posted, was modified by Author) |
| V 1381 problem with ObjectComm-multi rosaluna [2004/09/15-19:49] |
| Response : Edit : Delete |
|
I have a problem with ObjectComm-multi. I have compiled and executed sucessfully ObjectComm-multi/RP3. I have compiled sucessfully ObjectComm-multi/RP2 but I have problem with the execution: after that the connection is established with all gates, my aibo "dies"! |
| V 1379 Debian - installation problem mooti [2004/09/14-07:12] |
| Response : Edit : Delete |
|
I
am running a current version of debian unstable. Here is the
errors/warnings that I get while compiling binutils-2.14. Can anyone give me some pointers as to how to fix it? /bin/sh ./libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I../../binutils-2. 14/bfd -I. -D_GNU _SOURCE -DTRAD_CORE -I. -I../../binutils-2.14/bfd -I../../binutils-2. 14/bfd/../include - I../../binutils-2.14/bfd/../intl -I../intl -W -Wall -Wstrict-prototypes -Wmissing-prototype s -g -O2 -c ../../binutils-2.14/bfd/archive.c gcc -DHAVE_CONFIG_H -I. -I../../binutils-2.14/bfd -I. -D_GNU_SOURCE -DTRAD_CORE -I. -I../../b inutils-2.14/bfd -I../../binutils-2.14/bfd/../include -I../../binutils-2. 14/bfd/../intl -I../ intl -W -Wall -Wstrict-prototypes -Wmissing-prototypes -g -O2 -c ../.. /binutils-2.14/bfd/arch ive.c -o archive.o In file included from ../../binutils-2.14/bfd/archive.c:132: ../../binutils-2.14/bfd/libbfd.h:483: warning: `struct sec' declared inside parameter list ../../binutils-2.14/bfd/libbfd.h:483: warning: its scope is only this definition or declarati on, which is probably not what you want ../../binutils-2.14/bfd/archive.c: In function `get_extended_arelt_filename': ../../binutils-2.14/bfd/archive.c:328: warning: implicit declaration of function `strtol' ../../binutils-2.14/bfd/archive.c: In function `_bfd_generic_read_ar_hdr_mag': ../../binutils-2.14/bfd/archive.c:417: warning: implicit declaration of function `atoi' ../../binutils-2.14/bfd/archive.c: In function `do_slurp_coff_armap': ../../binutils-2.14/bfd/archive.c:817: warning: assignment from incompatible pointer type ../../binutils-2.14/bfd/archive.c:834: warning: assignment from incompatible pointer type ../../binutils-2.14/bfd/archive.c: In function `_bfd_write_archive_contents': ../../binutils-2.14/bfd/archive.c:1823: error: `PACKAGE' undeclared (first use in this functi on) ../../binutils-2.14/bfd/archive.c:1823: error: (Each undeclared identifier is reported only o nce ../../binutils-2.14/bfd/archive.c:1823: error: for each function it appears in.) ../../binutils-2.14/bfd/archive.c: In function `_bfd_compute_and_write_armap': ../../binutils-2.14/bfd/archive.c:1892: warning: implicit declaration of function `free' ../../binutils-2.14/bfd/archive.c: In function `_bfd_archive_bsd_update_armap_timestamp': ../../binutils-2.14/bfd/archive.c:2090: error: `PACKAGE' undeclared (first use in this functi on) |
| V 1373 Where to download OPEN-R 1.1.3-R2 aibo04 [2004/09/10-22:20] |
| | Response : Edit : Delete |
|
| Does
anyone know where to download the version 1.1.3-R2 of the OPEN-R? | | /Regards | |
| |- 1375 RE:Where to download OPEN-R 1.1.3-R2 AiboPet [2004/09/13-23:47] |
| | Response : HomePage : Edit : Delete |
|
| >Does
anyone know where to download the version 1.1.3-R2 of the OPEN-R? | | The current version 1.1.5-r3 is on the downloads page. | You may find older versions by editing the URL, however it appears that very | old versions like 1.1.3 are no longer on the website. | [FWIW: I kept an old backup] | | Why do you need the "1.1.3" version ? Do you need it all or just the memory | stick runtime ? | |
| +- 1378 RE:Where to download OPEN-R 1.1.3-R2 openrsupport@(Administrator) [2004/09/14-02:15] |
| Response : Edit : Delete |
|
I'm
sorry, OPEN-R 1.1.3-R2 is no longer available. Please use the latest OPEN-R SDK (OPEN-R SDK 1.1.5-r3). The redistribution of the OPEN-R SDK is not allowed. (See <END-USER LICENSE AGREEMENT> in the download page) AIBO SDE Support |
| V 1347 NaN and Inf crashes ejt [2004/09/04-02:07] |
| | Response : HomePage : Email : Edit : Delete |
|
| I'm
finding it quite frustrating that any operations involving NaN or Inf | floating point values will simply crash instead of giving the usual results | you'd get if the code were running on a desktop machine. It's not the initial | divide-by-zero or whatever which will cause the crash, it's only when the | resulting value is then used in any further operations (including output) which | causes the crash. | | What's even more frustrating is that even simply testing for these values using | isinf()/isnan()/finite() will also crash when they are passed NaN or Inf instead | of returning the appropriate result. So I have to write my own version which | tests the bit pattern directly, avoiding any floating point operations with the | potentially bad values. | | Since the admins don't seem to be responding to posts anymore, has anyone out | there gotten NaN/Inf results which *didn't* crash? Perhaps there's something | I'm doing... | | (Also still waiting on fstop vs. focal length clarification, and the ers-7 leg pose) | |
| +- 1368 RE:NaN and Inf crashes openrsupport@(Administrator) [2004/09/09-08:51] |
| | Response : Edit : Delete |
|
| The
Crash sample shows cases which can cause floating point exceptions. | Note: isinf()/isnan()/finite() are not OPEN-R functions. You can see the | mipsel header file and source code to find out why they behave the way they do. | | AIBO SDE Support | |
| +- 1371 RE:NaN and Inf crashes ejt [2004/09/10-07:03] |
| Response : HomePage : Email : Edit : Delete |
|
>The
Crash sample shows cases which can cause floating point exceptions. We agree that this causes a crash. My point is that it *should not* cause a crash. This behavior is not consistant with expectations of the IEEE standard, and would be considered by many (such as me) to be a bug for cases other than a SNaN (Signalling NaN). Inf and NaN have well defined operations, and should be implemented. http://www.google.com/search?q=IEEE+floating+point >Note: isinf()/isnan()/finite() are not OPEN-R functions. You can see the >mipsel header file and source code to find out why they behave the way they do. Although this is pushing the limits of my low-level knowledge, I think this is an issue with the operating system's handling of floating point CPU exceptions, not the library code. Instead of giving up and crashing at the first sign of trouble, can't it just ignore and continue like modern computers do? Try it: #include <stdio.h> int main() { printf("%g\n",1.0/0.0); } /* displays 'inf', not core dump */ |
| V 1302 TCP receiving rate NWCIL [2004/08/13-03:54] |
| | Response : Edit : Delete |
|
| Hello
- I'm trying to increase the frequency at which OPEN-R polls a TCP
socket | for new messages. Right now, it appears to do this every ~200ms. TCP messages | can be sent much faster than that, and I would like to process them at a higher | rate (I'm sending joint positions from a different computer, and getting updates | at 200ms results in very jerky motion). Does anyone have an idea for where I | should look for the code responsible for polling TCP sockets? | | thanks, | Bryan Johnson | NorthWest Computational Intelligence Lab | |
| +- 1367 RE:TCP receiving rate openrsupport@(Administrator) [2004/09/09-08:00] |
| | Response : Edit : Delete |
|
| TCP
polling can be increased by stopping other processes that are running on | AIBO. You will have a high-speed network box, but you may loose functionality | which makes AIBO a "robot". | | AIBO SDE Support | |
| +- 1370 RE:TCP receiving rate NWCIL [2004/09/09-12:12] |
| Response : Edit : Delete |
|
Thanks
for your reply. I had little success with TCP, so I transferred all my communications to UDP, which is much MUCH faster. This is an infinitely better protocol for real-time control of the robot. |
| V 1323 ERS-7 documentation diagrams ejt [2004/08/27-05:06] |
| | Response : HomePage : Email : Edit : Delete |
|
| What
are the joint angles of the knees shown on page number 3 of
ModelInformation_7? (that's physical | page page 5) | The leg blow-ups on page number 4 have prominent 30 degree lines, without labeling what exactly | those lines are in reference to. I assume they show the vertical axis of the joint? However, if the | dashed line on page number 5 indicates the vertical axis, then this line is 35 degrees (measured | precisely by blowing up the diagram in Adobe Illustrator). Further, the 210 and 220 models are shown | with knees at 35 degrees, so that would be more consistent. | | We're trying to get precise measurements of points on the legs so we can do manipulations with them. | But being off by 5 degrees would add error without being visually obvious :( | | Feedback would be greatly appreciated! | | Thanks! | -ethan | |
| +- 1331 RE:ERS-7 documentation diagrams garfae [2004/08/30-21:34] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
think the schematics on page 5 are not drawn to proper scale, i.e.
they're | just to give you min and max joint angle values, as well as their direction. You | should trust page 3+4, titled "External Measurements" (whereas page 5+ is | "Hardware Limitations of Joint Motion")... | | -Sp | |
| +- 1338 RE:ERS-7 documentation diagrams ejt [2004/09/01-00:12] |
| | Response : HomePage : Email : Edit : Delete |
|
| The
extra trick is that the legs from 3 & 5 line up perfectly, where as
the leg from page 4 doesn't quite | really line up very well - it's a slightly different shape as well as a different orientation. So even if the | 0-mark on page 5 isn't accurately drawn, there's still evidence that page 4 may be a different position. | Even if that is shown at 30 degree position, unfortunately it's the diagram on page 3 that I really care | about the position of. | |
| +- 1369 RE:ERS-7 documentation diagrams openrsupport@(Administrator) [2004/09/09-10:27] |
| Response : Edit : Delete |
|
The
diagrams are not drawn precisely, so taking measurements manually will
not be accurate. The heights of the knees on page 3 are the same as page 4, so the angles of the knees are the same. Thank you, AIBO SDE Support |
| V 1300 f-stop vs. focal length ejt [2004/08/11-03:01] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
just want to verify - from the model information for the ERS-7 | "Lens: F 2.8, f = 3.27mm" | This means f-stop is 2.8, and focal length is 3.27 mm? | Thanks! | |
| +- 1366 RE:f-stop vs. focal length openrsupport@(Administrator) [2004/09/09-07:49] |
| Response : Edit : Delete |
|
Yes, you are correct. F is F-Stop (F value or F number) - the amount of light which passes through the lens f is focal distance AIBO SDE Support |
| V 1365 OPEN-R SDK 1.1.5-r3 has been released openrsupport@(Administrator) [2004/09/08-01:10] |
| Response : Edit : Delete |
|
* OPEN-R SDK 1.1.5-r3 has been released. EEPROM protection after calling OPENR::Fatal() TCPGateway support for RoboCup * OPEN_R_SDK-sample-1.1.5-r2.tar.gz has been released. ObjectComm-multi and NTP sample programs have been added AIBO SDE Support |
| V 1320 AIBO navigation project --- camera questions Ethanlien [2004/08/25-19:21] |
| | Response : Email : Edit : Delete |
|
| Hi all! I am new to AIBO and its capabilities! | | I am a third year university student about to start my masters project soon on | a home robot that can navigate by itself through recongnition and learning via | its camera. | | I have been suggested to look into the SONY AIBO as the basis for my project | with intention to buy one =) | | In order to track where it is, we need to extract teh image files from its | camer. Can we do this on the AIBO? | | If so perhaps tell me the steps to do so? =)) Much appreciated | | E | |
| +- 1326 RE:AIBO navigation project --- camera questions AiboPet [2004/08/28-00:42] |
| | Response : HomePage : Edit : Delete |
|
| >In
order to track where it is, we need to extract teh image files from its | >camer. Can we do this on the AIBO? | | There are several ways to do this, and several different tools (programming | APIs). Several questions for you (to figure out the best recommendation): | | Are you planning on having a WiFi connection to a PC to do the image | processing or do the image processing on the robot ? | Have you picked the model yet (ERS-210, ERS-220 or the current ERS-7 all have | programmability and WiFi LAN capability) | | Do you have C/C++ experience ? | Would you rather write the program on the robot or on a PC ? | |
| +- 1341 RE:AIBO navigation project --- camera questions Ethanlien [2004/09/03-00:25] |
| | Response : Email : Edit : Delete |
|
| ERS-7,
is that the best for it? | | I have a 33% discount due to being a research programme. | | I have some C++ and quite a lot of theoretical knoledge, but low practice. | | I guess edge/corner detection on PC... | | Weisi | |
| +- 1351 RE:AIBO navigation project --- camera questions AiboPet [2004/09/04-03:31] |
| | Response : HomePage : Edit : Delete |
|
| >ERS-7,
is that the best for it? | | Without knowing more about your project -- probably yes. | | I am guessing that you may want to use the Remote Framework (image capture, | walking, pattern recognition all built under PC WiFi control) | You can write all your code and debug it on a PC (eg: using Visual C++) and | avoid many of the hassles of writing and debugging code running on the robot | itself. | | ---- | Technically The ERS-7 is the best robot for almost everything. | | The one caveat is that if you want to use the OpenR SDK, and have your AIBO | walk around normally (not on his knees), you have to write your own walking | engine. In that case the ERS-210/220 is a better solution. | | The ERS-210/220 has a sample walk engine provided by Sony (OMWares, and the | Soccer Lion sample). They haven't ported it to the ERS-7 (yet, and probably | never unless we keep pressuring them) | | See message #1297 on this BBS for more info | |
| +- 1356 RE:AIBO navigation project --- camera questions garfae [2004/09/07-01:22] |
| | Response : HomePage : Email : Edit : Delete |
|
| >The
ERS-210/220 has a sample walk engine provided by Sony (OMWares, and the | >Soccer Lion sample). They haven't ported it to the ERS-7 (yet, and probably | >never unless we keep pressuring them) | | Ummmmmm...! Where do we have to sign? | | Then again, you can always do MTN playback on the robot if you have a decent | motion definition, right? | | -Sp | |
| +- 1358 RE:AIBO navigation project --- camera questions AiboPet [2004/09/07-07:15] |
| | Response : HomePage : Edit : Delete |
|
| >Then
again, you can always do MTN playback on the robot if you have a decent | >motion definition, right? | | Not really. | If you try the MTN playback sample (the old ERS-210 ones you mention in | another thread) and compare it with the algorithmic one (OMWares/SoccerLion | 210/220 only), I think you will find a big difference. | Also it handles transitions and parameterization of the walking. | | If important to you, please post a request (either in a separate thread, or | reply to #1297) | NOTE: I am only asking for parity for the ERS-210 SoccerLion sample (not | asking for all the ERS-7 middleware) | |
| +- 1361 RE:AIBO navigation project --- camera questions garfae [2004/09/07-18:55] |
| | Response : HomePage : Email : Edit : Delete |
|
| >NOTE:
I am only asking for parity for the ERS-210 SoccerLion sample (not | >asking for all the ERS-7 middleware) | | Well, would it hurt them to provide the whole of it, anyway? It's not like | people will be making money on their backs just because they release the | algorithms they developped. | | Perhaps they're holding back to increase the challenge in the RoboCup contest. I | don't know... | | -Sp | |
| +- 1363 Missing OMWare (was re: AIBO navigation project) AiboPet [2004/09/07-23:28] |
| | Response : HomePage : Edit : Delete |
|
| >Well,
would it hurt them to provide the whole of it, anyway? | | Long story. But asking for *all* of the Middleware is a sure way for them to | say no (see other post). The Middleware and App layers of their commercial | software are treated like "the family jewels". | http://www.aibohack.com/openr_sdk - see diagram at bottom | | I disagree with this decision since it creates separate and unequal program | choices (eg: don't wait for OpenR addins to RCODE or RFW programs either ;-< ) | | Feel free to ask for anything else you may want, but please separate it from | this simple request (and IMHO a very reasonable one) | | ----- | The OMWare module in question, was only released in binary form (similar to | the AiboLife1 walking). Many improvements have been made for the ERS-7, some | of which are licensed from other companies (evolution.com for the pattern | recognition) | | I'd be happy with the Leg module from OMWare working on the ERS-7, binary | only, just like the ERS-210. They have done this before, so it is not | unreasonable to expect it. | | ------- | >Perhaps they're holding back to increase the challenge in the RoboCup contest. | Most of the software in question is not what a RoboCup dog needs. | eg: safe walking, pattern recognition, high level color tracking, voice | recognition, sound controller detection, ... (ie. features you do get in RCODE | or RFW but not in the OpenR SDK) | | These features are very important for creating "entertainment robot | personalities". That's why RCODE (or RFW) is so much better for that kind of | software. Unfortunately, I believe Sony wants to keep it that way. | |
| +- 1364 RE:Missing OMWare (was re: AIBO navigation project) garfae [2004/09/08-00:45] |
| Response : HomePage : Email : Edit : Delete |
|
>Long
story. But asking for *all* of the Middleware is a sure way for them to >say no (see other post). Agreed. I won't be asking as I have no need for it yet. But maybe someday... >I disagree with this decision since it creates separate and unequal program >choices (eg: don't wait for OpenR addins to RCODE or RFW programs either ;-< ) I suppose you can find strong arguments supporting both sides... Still, it's their code to dispose of as they please, so all we can do is keep asking @|:-) >The OMWare module in question, was only released in binary form (similar to >the AiboLife1 walking) [...] Thanks for all the info, and for the link to aibohack -- very helpful! My boss has told me about the owner of that site, apparently they've helped us last year with custom MTN files (those I mentioned in another thread). -Sp |
| V 1362 MotionEditor feature request garfae [2004/09/07-19:03] |
| Response : HomePage : Email : Edit : Delete |
|
Regarding the new & shiny Motion Editor software: The application is quite well-done, if not particularly user-friendly; however, we have found it lacking a feature which would be extremely helpful to us, which is this: The software allows the user to manually force keyframes into the motion; extending this, we would need an *automatic* keyframe generator, e.g. a functionality (button,option,...) which would automatically insert a keyframe every n frames. (Possibly only when exporting as MTN, even, since this is where the feature is really useful.) Why do we need this? Because we have an extremely simple custom MTN playback routine, and it doesn't work that well with large steps between keyframes (typically, we used to have keyframes every 8ms, i.e. every 1 frame, in our old MTN files). Also, the spline interpolation feature can in this way be capitalized upon, in more than just showing the animation in Motion Editor... Thanks, -Sp |
| V 1285 OMWares modules theoldlr [2004/08/04-00:32] |
| | Response : Email : Edit : Delete |
|
| Could the OMWares modules please be ported to the ERS-7? | | I would much appreciate it. | | Thanks! | |
| +- 1295 RE:OMWares modules openrsupport@(Administrator) [2004/08/11-02:12] |
| | Response : Edit : Delete |
|
| I'm
sorry, but we cannot release the OMWares module for the ERS-7. The
ERS-7's | walking modules are very complex and tightly connected to other commercial | modules. This allows AIBO (ERS-7) to have more animal like motions and to make | natural transitions between the walking posture and other postures. | | If you want to use the commercial based walking module, please use AIBO Remote | Framework. It will allow you to use very powerful commercial based functions. | | Thank you, | | AIBO SDE Support | |
| +- 1297 RE:OMWares modules (restating request) AiboPet [2004/08/11-02:47] |
| | Response : HomePage : Edit : Delete |
|
| >I'm
sorry, but we cannot release the OMWares module for the ERS-7... | | It would be nice to get a release of all the middle-ware of the ERS-7 AiboMind | stick with all the new features - but I don't believe that is what is being | asked for. | | ---- | restating the request: | | What is needed is a port of the ERS-2x0 OMWares module - in particular the leg | walking module. No new functionality, just the same functionality already | available to OpenR SDK programmers who have an ERS-210 or ERS-220. | | Currently there is no "Soccer Lion" equivalent for the ERS-7, and no way to | make the ERS-7 walk. | [unless you use something like the CMU Knee walk which is hard on the dog] | | Having the dog walk should not be limited to RFW and RCODE apps -- OpenR SDK | apps on the ERS-7 should have this capability too. | | This is a gaping hole IMHO | | Thanks for your re-consideration | |
| +- 1360 RE:OMWares modules (restating request) garfae [2004/09/07-18:53] |
| Response : HomePage : Email : Edit : Delete |
|
>Having
the dog walk should not be limited to RFW and RCODE apps -- OpenR SDK >apps on the ERS-7 should have this capability too. > >This is a gaping hole IMHO I agree with this. It would be extremely helpful for all developpers (I don't have commercial development in mind, but rather academic research) to have this module readily available. Of course it should be possible to re-discover it, with proper time and research funding, but it would be a waste of effort which should rightly be put to some other, more fruitful, use intead... So we join the motion: please re-consider supporting the ERS-7 in the same way you used to support ERS-210. Thanks, -Sp |
| V 1354 Has the memory stick virus ? robottj [2004/09/06-21:33] |
| | Response : Edit : Delete |
|
| Hi, | in my programming memory stick there is a file named"DR.BIN". It's size is | 0,but I can't delete it.Is it possible that my memory stick has virus ? If | so ,what can I do for it ? | Thanks for any help! | |
| +- 1357 RE:Has the memory stick virus ? ejt [2004/09/07-04:58] |
| | Response : HomePage : Email : Edit : Delete |
|
| Pretty
doubtful - I'd say it's much more likely your memory stick has some
corruption, from say, taking | it out of a writer while it's being accessed (or such). | | If you have a linux machine, it should have something like /sbin/fsck.msdos which can usually fix this | kind of thing. I'd bet windows also has some kind of disk repair if you go to the properties for the | drive. | |
| +- 1359 RE:Has the memory stick virus ? AiboPet [2004/09/07-07:20] |
| Response : HomePage : Edit : Delete |
|
FWIW:
virus infections have been reported before. It is possible that your COMPUTER has a virus. Inserting the memory stick into a virus infected computer may cause the virus (on your computer) to try to infect the memory stick The result is a damaged stick image (the PC virus won't hurt AIBO, he just won't boot). If you suspect a virus, I strongly recommend you update your virus definitions and run a full system virus scan. Also scan the memory sticks (ie. if you clean up your PC, but have an old infected memory stick, it can re-infect your PC -- AIBO is safe, but the stick can propagate PC viruses) As mentioned, it is possible that the memory stick FAT format is a little screwed up, for other reasons. ===== My advice is to try a *QUICK* format using a WinXP, 2K or NT machine. "FORMAT /Q ?:" where ? is the drive letter. WARNING: ONLY USE A SONY BRAND STICK READER WARNING: Do not full format (AKA unconditional format) WARNING: Do not format using Win98/WinME etc. The WinXP version of format is safer and better. This configuration has fixed almost all stick problems I've seen, except if the low level copy protection is gone. |
| V 1355 MTN files for ERS-7 :: walk, crawl, etc. garfae [2004/09/07-02:24] |
| Response : HomePage : Email : Edit : Delete |
|
Last
year, when we were modelling the ERS-210, somebody was kind enough to supply us with custom MTN files for walking and crawling. So, I was wondering if this time again, anyone here could help us with this? We need a walking MTN in particular, but any motion expression MTN's would help (crawl, get down/get up, jump maybe? <-- we're using a simulated model only so it's quite safe). (Certainly, we're not the only ones needing a walking motion to test the robot, and while we could develop something on our own using the Motion Editor, we figured we'd ask first, so as not to re-invent the wheel...) -- EDIT --- We managed to adapt old ERS-210 MTN's to new ERS-7 definition (we did try that before, but the robot model was incorrect and therefore the motion seemed wrong). But still, if anyone has additional MTN's, they're very welcome! We can also share ours, of course. ----------- Thanks in advance! -Sp (The article that 2004/09/07-01:11 posted, was modified by Author) |
| V 1353 Author DELETE ***** [2004/09/06-20:28] |
| Response : Edit : Delete |
|
(The
article that 2004/09/06-20:08 posted, was modified by Author) |
| V 1332 ERS-7 chest distance sensor garfae [2004/08/30-23:02] |
| | Response : HomePage : Email : Edit : Delete |
|
| I have a question regarding the ERS-7 chest distance sensor: | | The range specified in the reference is (10,90) cm; does that mean that | anything closer than 10 cm will have the same sensor reading (i.e. 10 cm)? | | Also, we don't have a real ERS-7 underhand, but if any owner is willing to do | the experiment, it will be very helpful. The basic test should be done by | reading values from the distance sensor while placing a sheet of paper at | predetermined distance (10,20,30...cm) and writing the value read for each | distance (maybe even with an error margin) in a table. (Same with the two other | distance sensors.) | | Thanks in advance to anyone willing to help, | -Sp | (The article that 2004/08/30-22:21 posted, was modified by Author) | |
| +- 1337 RE:ERS-7 chest distance sensor AiboPet [2004/08/31-00:51] |
| | Response : HomePage : Edit : Delete |
|
| >I
have a question regarding the ERS-7 chest distance sensor: | > | >The range specified in the reference is (10,90) cm; does that mean that | >anything closer than 10 cm will have the same sensor reading (i.e. 10 cm)? | Yup. | | I believe they use a Sharp GP2D02 or similar device. | | I haven't done a detailed set of measurements as you suggest, but I find the | distance sensor does a pretty good job when up against a wall (big flat object | at equal distance), but is much less accurate with smaller objects in front of | his nose or at an angle. | What kind of objects are you trying to detect ? (ie. I suspect not flat pieces | of paper) | |
| +- 1352 RE:ERS-7 chest distance sensor garfae [2004/09/06-20:06] |
| Response : HomePage : Email : Edit : Delete |
|
>What
kind of objects are you trying to detect ? (ie. I suspect not flat
pieces >of paper) Currently I am not trying to detect anything, just to model the robot properly in our modelling software (Webots). Thanks anyway! -Sp |
| V 1340 Data sended by subject differ from data received by observer aiboKang [2004/09/02-12:17] |
| | Response : Edit : Delete |
|
| Hi, | | I'm trying to build a program playing robocup game. Behavior and Motion are | two of the modules in my program. Data should be transmitted from Behavior to | Motion.But I found that the data sended by Behavior differed from the data | received by Motion. Here are the codes used in my program: | | 1. send data in Behavior: | | if (subject[sbjCmuSetMotion]->IsReady()) { | subject[sbjCmuSetMotion]->SetData(&mc, sizeof(mc)); | subject[sbjCmuSetMotion]->NotifyObservers(); | } | | 2. receive data in Motion: | | memcpy(&cmd, event.Data(0), sizeof(cmd)); | observer[event.ObsIndex()]-> AssertReady(event.SenderID()); | | I used "OSYSPRINT()" to show the values of 'cmd' and 'mc' respectively in | Motion and Behavior,which should be the same , but unexpectedly they were | different seeing from wconsole. Here are the values of 'cmd' and 'mc' showed | on wconsole: | | cmd.motion_cmd=1,cmd.head_cmd=200,cmd.vx=0.000000,cmd.vy=192.233010,cmd.va=0.00 | 0000,cmd.head_pan=-0.165957,cmd.head_tilt=0.000000 | | mc.motion_cmd=200,mc.head_cmd=1, mc.head_pan=0.099029, mc.head_tilt=-0.165957, | mc.vx=192.233010, mc.vy=0.000000, mc.va=0.065359 | | what kind of cause can it be? How can I resolve it? | | Any tip would be appreciated. | | kandar | (The article that 2004/09/02-12:05 posted, was modified by Author) | |
| +- 1350 RE:Data sended by subject differ from data received by observer AiboPet [2004/09/04-03:22] |
| Response : HomePage : Edit : Delete |
|
>
...But I found that the data sended by Behavior differed from the data >received by Motion. Here are the codes used in my program: >cmd.motion_cmd=1,cmd.head_cmd=200,cmd.vx=0.000000,cmd.vy=192.233010,cmd.va=0.0 0 >0000,cmd.head_pan=-0.165957,cmd.head_tilt=0.000000 > >mc.motion_cmd=200,mc.head_cmd=1, mc.head_pan=0.099029, mc.head_tilt=-0.165957, >mc.vx=192.233010, mc.vy=0.000000, mc.va=0.065359 I assume both 'cmd' and 'mc' are MotionCommand types. The fact that the 'motion_cmd' and 'head_cmd' fields are swapped is suspicious [assuming no typos]. Otherwise I suspect an alignment error, or other data corruption problem. Debug suggestion: print out the data sent and data received in hex bytes instead, that may be more illuminating. There is also the possibility that the 'mc' data is coming from some other source (some other object sending MotionCommands) |
| V 1339 Remote Framework --- VPRQueryMode Fairlawn07 [2004/09/01-05:33] |
| | Response : Email : Edit : Delete |
|
| I
have problem to understand how to use VPRQueryModel function. | I am retrieving the results with the WM_VAIBO_INPUT_SEMANTICS Message (labelID | = SEMID_VPR_RESULT), but cannot make any sense of the results. | | Feedback would be greatly appreciated! | |
| |- 1344 RE:Remote Framework --- VPRQueryMode TodboT [2004/09/03-09:52] |
| | Response : Edit : Delete |
|
| I
think VPRQueryModel is used to check if a particular model ID already
has a | trained image. You pass VPRQueryModel the ID & sub ID you want to check, and | the result is returned in a SEMID_VPR_RESULT semantics message. subInfo[2] will | be 0xFFFFFFFF if the model doesn't exit, or it will be the model sub ID if it | does exist. | |
| +- 1349 RE:Remote Framework --- VPRQueryMode AiboPet [2004/09/04-03:08] |
| Response : HomePage : Edit : Delete |
|
>I
have problem to understand how to use VPRQueryModel function.... The VPR interface is minimally documented. As mentioned the VPRQueryModel / SEMID_VPR_RESULT will give you information about what is trained for one model (ie. does not perform pattern match) You pass in the model number. The SEMID_VPR_RESULT returns -1 if nothing trained, otherwise the highest sub-model number trained. In theory you could use that to determine the next submodel to train (using EVPRPOLICY0) FWIW: I use a single sub-model for each model (POLICY1), so I don't use it. ----- BTW: To detect a trained model, you need to look for the SEMID_EPTN notification, which is triggered when AIBO detects one of the trained patterns. This will give you the model and sub-model index of the best matched pattern, as well as the rotation and quality of the match |
| V 1346 How can I get the IP Address msgid [2004/09/03-19:45] |
| | Response : Edit : Delete |
|
| I'm
implementing a protocol to cumincate between two or more AIBO. I use a
file | that contain the IP address of the other AIBO, but I don't know how can I | retreive the IP address of the AIBO that is running the protocol... how can I do ? | |
| +- 1348 RE:How can I get the IP Address AiboPet [2004/09/04-03:03] |
| Response : HomePage : Edit : Delete |
|
> ... know how can I >retreive the IP address of the AIBO that is running the protocol... how can I do ? Take a look at the "ERA201D1Info" sample. It extracts a number of interesting statistics from the WiFi LAN card, including the ETHER_IP address |
| V 1293 about the stereo microphones totoduke [2004/08/10-01:16] |
| | Response : Email : Edit : Delete |
|
| Hi, | | Is there any possible that the aibo can tell the direction of from which a | sound is? | | Thanks a lot. | |
| +- 1296 RE:about the stereo microphones openrsupport@(Administrator) [2004/08/11-02:17] |
| | Response : Edit : Delete |
|
| If
you have an ABIO ERS-7, AIBO Remote Framework can give you the direction | towards a sound. | | Please see "Semantics.txt" in the AIBO Remote Framework distribution. | | Thank you, | | AIBO SDE Support | | + from Audio Recoginizer | | SEMID_VOICE ( 0x0301 ) // Voice Recognition | // SubInfo[0] Indicate recognized voice | --> See InputSemantics SubInfo list below | // SubInfo[1] The angle where the voice comes from | (degrees; left is positive) | // SubInfo[2] Speaker recognition | --> See InputSemantics SubInfo list below | | SEMID_LOUD ( 0x0701 ) // Detect a loud noise | // SubInfo[0] indicate sound type --> | See InputSemantics SubInfo list below | // SubInfo[1] The angle where the voice comes from | (degrees; left is positive) | | + Audio Recognizer : SEMID_LOUD (Sound detect) SubInfo[0] | M MASK_LOUD_ELEMENT 0xffffffff LOUD /* Sound attribute */ | S LOUD_LONG 0x00000001 MASK_LOUD_ELEMENT /* long sound */ | S LOUD_SHORT 0x00000002 MASK_LOUD_ELEMENT /* short sound */ | S LOUD_PALSE 0x00000003 MASK_LOUD_ELEMENT /* palse sound */ | S LOUD_SH 0x00000004 MASK_LOUD_ELEMENT /* internal use */ | S LOUD_GARBAGE 0x00000005 MASK_LOUD_ELEMENT /* noise */ | S LOUD_CANCEL_SIT 0x00000006 MASK_LOUD_ELEMENT /* transition to sit | posture noise */ | S LOUD_CANCEL_STAND 0x00000007 MASK_LOUD_ELEMENT /* transition to stand | posture noise */ | S LOUD_CANCEL_WALK 0x00000008 MASK_LOUD_ELEMENT /* walk noise */ | | + Audio Recognizer : SEMID_LOUD (Detect sound) SubInfo[1] | M MASK_ANGLE_ELEMENT 0xffffffff VOICE LOUD | /* Direction of the sound */ | |
| +- 1298 RE:about the stereo microphones AiboPet [2004/08/11-02:51] |
| | Response : HomePage : Edit : Delete |
|
| Also
an RCODE alternative for ERS-210/220 AIBOs, running RcodePlus | | AP_Voice_Horz gives a rough horizontal position of the last voice command | AP_Loud_Horz gives a rough horizontal position of the last loud noise | | http://www.aibohack.com/rcode/rcp_sound.htm | | Not available for ERS-7 RCODE | | For OpenR SDK programming, you can guestimate the position based on the | difference in volume between the two ear mics (no voice commands however) | |
| +- 1303 RE:about the stereo microphones totoduke [2004/08/13-06:17] |
| | Response : Email : Edit : Delete |
|
| >Also
an RCODE alternative for ERS-210/220 AIBOs, running RcodePlus | > | >AP_Voice_Horz gives a rough horizontal position of the last voice command | >AP_Loud_Horz gives a rough horizontal position of the last loud noise | > | >http://www.aibohack.com/rcode/rcp_sound.htm | > | >Not available for ERS-7 RCODE | > | >For OpenR SDK programming, you can guestimate the position based on the | >difference in volume between the two ear mics (no voice commands however) | | Hi AiboPet, | | Thank you for your reply. That is very helpful. | But I still have a question: How can I get the signals from the two ear | microphones? Now I have recorded the sound and read it from Matlab. But there | was only one signal!! Can I set something that I can record two different | signals from the two microphones? | | Thank you very much!! | |
| +- 1304 RE:about the stereo microphones AiboPet [2004/08/14-02:26] |
| | Response : HomePage : Edit : Delete |
|
| >But
I still have a question: How can I get the signals from the two ear | >microphones? Now I have recorded the sound and read it from Matlab. But there | >was only one signal!! | | The lower-level system (OpenR SDK, and RCodePlus telemetry) records the audio | as 16-bit stereo samples (usually 16kHz IIRC) | | How are you getting it into Matlab ? | | I know of two Matlab interfaces: | Udi's "ABO" toolbox for the 2x0 that uses RCodePlus (stereo data is available, | unless the MatLab wrappers strip it to mono) | http://www.benreuven.com/micresponse.htm | | Also the Tekkotsu interfaces that uses their own protocol. | |
| +- 1305 RE:about the stereo microphones AiboPet [2004/08/14-02:29] |
| | Response : HomePage : Edit : Delete |
|
| >Udi's
"ABO" toolbox for the 2x0 that uses RCodePlus (stereo data is available, | >unless the MatLab wrappers strip it to mono) | >http://www.benreuven.com/micresponse.htm | | (according to the schematic diagram) | Looks like the ABO interface is only exposing the "left" microphone. The | MatLab glue will have to be tweeked to get both channels. | |
| +- 1309 RE:about the stereo microphones totoduke [2004/08/17-04:44] |
| | Response : Email : Edit : Delete |
|
| >>Udi's
"ABO" toolbox for the 2x0 that uses RCodePlus (stereo data is | available, | >>unless the MatLab wrappers strip it to mono) | >>http://www.benreuven.com/micresponse.htm | > | >(according to the schematic diagram) | >Looks like the ABO interface is only exposing the "left" microphone. The | >MatLab glue will have to be tweeked to get both channels. | | I am sorry that I didn't explain my problem clearly. My robot is ERS-7 and I | am sure if I can use the Rcode in this type of Robots. And Now I only use the | sample "SoundRec" provided by sony company to record outer sound. It seemed | that I could only get one signal. Can I control some variables in OPENR to get | both signals from the two microphones. And even more, can I get four signals | from the two pairs of microphones simultaneously? | |
| |- 1313 RE:about the stereo microphones AiboPet [2004/08/19-00:58] |
| | | Response : HomePage : Edit : Delete |
|
| | >I
am sorry that I didn't explain my problem clearly. My robot is ERS-7
and I | | >am sure if I can use the Rcode in this type of Robots. | | | | Yes you can, but RCODE for the ERS-7 doesn't have that kind of sound | | capabilities. | | | | > And Now I only use the | | >sample "SoundRec" provided by sony company to record outer sound. | | | | That's the sample to use. | | I think there is some confusion over the number of channels you are expecting | | (see below) | | | | ====== | | > can I get four signals | | >from the two pairs of microphones simultaneously? | | | | Not AFAIK. | | | | ----- | | For the ERS-210/220: | | | | Only one microphone "mode" can be set at a time (either UNIdirectional or | | OMNIdirectional). | | Set the mode you are interested in, and then you can get a stereo signal (2 | | channels of 16bit audio samples) | | | | I don't know if the hardware is actually 2 or 4 separate microphones -- but | | AFAIK there are only 2 available at a time for the software. | | | | For the ERS-7: | | I don't know if they support both modes or not. | | The SoundRec sample doesn't set the mode for the ERS-7 model. | | |
| | +- 1314 RE:about the stereo microphones AiboPet [2004/08/19-01:27] |
| | Response : HomePage : Edit : Delete |
|
| UPDATE: | testing with AiboScope7, as expected the OMNI and UNI feature does not work on | the ERS-7 | | The ERS-7 model specs say it doesn't support this feature or ALC | | (the ERS-2x0 does support OMNI/UNI) | |
| |- 1315 RE:about the stereo microphones TodboT [2004/08/19-12:29] |
| | | Response : Edit : Delete |
|
| | You
can only get 2 channels of sound (left and right). I think the
confusion is | | that the one signal you think you are getting is actually two that are | | interleaved. I think the format of the recorded sound is 16-bit 16-khz stereo. | | The first two bytes of the received sound data are for the left channel (least | | significant byte first), then the next two bytes are for the right channel. | | Then left, right, left and so on. | | | | Basic sound localization is pretty easy to do. Lookup "interaural time | | difference". All you really need to do is shift one of the channels until the | | two channels are correlated. How much you had to shift the channel corresponds | | to the angle of the sound. If there was no shift, then the sound is in front | | (or behind) aibo. If you had to shift one of the channels, then it came from | | the side. Since the audio is sampled at 16-khz, and aibo's head isn't very big, | | there's not a lot of resolution or accuracy, but it does work fairly well. | | |
| | +- 1318 RE:about the stereo microphones totoduke [2004/08/24-09:14] |
| | | Response : Email : Edit : Delete |
|
| | >You
can only get 2 channels of sound (left and right). I think the confusion | | is | | >that the one signal you think you are getting is actually two that are | | >interleaved. I think the format of the recorded sound is 16-bit 16-khz stereo. | | > The first two bytes of the received sound data are for the left channel | | (least | | >significant byte first), then the next two bytes are for the right channel. | | >Then left, right, left and so on. | | >Basic sound localization is pretty easy to do. Lookup "interaural time | | >difference". All you really need to do is shift one of the channels until the | | >two channels are correlated. How much you had to shift the channel corresponds | | >to the angle of the sound. If there was no shift, then the sound is in front | | >(or behind) aibo. If you had to shift one of the channels, then it came from | | >the side. Since the audio is sampled at 16-khz, and aibo's head isn't very | | big, | | >there's not a lot of resolution or accuracy, but it does work fairly well. | | | | Your response is very helpful. Yes, I found the acoustic signal I recorded | | using "SoundRec" was interweaved together. And when I read the signal out | | using matlab, there were indeed two signals. And I supposed that Matlab reads | | the *.wav files in according to what you said. And I used some simple way(very | | similar to what you said) and tried to get some certain angle from which the | | source came from. | | | | What I did was like that: | | 1) first I got the inner-product of the signals from the two channels | | 2) after finding the maximum point of the inner-product and since there were | | sampled in 16kHz, the time delay (t) of these two sigals could be derived by | | dividing the maximum point by the sampling frequency. | | 3)we knew the speed of the sound (v) and the distance (d) between the two ears | | and we could compute the angle (theta) using equation: cos(theta) = t*v/d. | | | | But the problem is: since v is very large (343m/s) and d is very small | | (approx. 0.115m), cos(theta) is much greater that 1. How can it be possible? | | And also, what does the quantized bit length (16 bit in ERS7)mean? Should I | | use it to compute the angle? Or is there any problem for matlab to read out | | data such like *.wav? | | | | |
| | +- 1345 RE:about the stereo microphones TodboT [2004/09/03-10:06] |
| | Response : Edit : Delete |
|
| >using
equation: cos(theta) = t*v/d. | > | >But the problem is: since v is very large (343m/s) and d is very small | >(approx. 0.115m), cos(theta) is much greater that 1. How can it be possible? | >And also, what does the quantized bit length (16 bit in ERS7)mean? Should I | >use it to compute the angle? Or is there any problem for matlab to read out | >data such like *.wav? | | You must be using the wrong value of t. If you shifted one of the channels by | 3, then t would be 3/16000. Using your equation: cos(theta) = | (3/16000)*343/0.115 = 0.559. | | 16 bit audio just means each sample of the sound is stored as a 16 bit value. | The more bits (and higher frequency), the more accurate the digitized audio will | be to the source. This shouldn't affect your calculations above. | |
| +- 1316 3rd option Remote FW Kuukai [2004/08/20-23:52] |
| Response : Edit : Delete |
|
How
about AIBO Remote FW? You have the 3rd option that you try to use AIBO Remote Framework. You can get AIBO's audio data by streaming (real time). + AudioGetOpen [Function] Start getting audio data [Return] AUDIOTP_NOERROR or Error code (see Error code list at the top of this text) [Description] Open the network port and start getting audio data. The audio data formats are ATP16K_XXX and ATP8K_XXX which are defined in VAIBODef.h [Usage] int CVAIBO::AudioGetOpen( int mode ) // audio data format (Defined in VAIBODef.h) + AudioGetData [Function] Get audio data [Return] AUDIOTP_NOERROR or Error code (see Error code list at the top of this text) [Description] Get audio data [Usage] int CVAIBO::AudioGetData( unsigned char *buff, // buffer that stores data (the buffer should be allocated) int size ) // data size (size is retrieved from the parameter with WM_AUDIOTP_UPDATE) // APP(AudioTP) command #define APPCMD_AUDIO_RECV_START 0x0040 // Receive data; AudioTP(AIBO- >PC) start #define APPCMD_AUDIO_RECV_STOP 0x0041 // Receive data; AudioTP(AIBO- >PC) stop #define APPCMD_AUDIO_RECV_MODE 0x0042 // Receive data; AudioTP(AIBO- >PC) change data mode #define APPCMD_AUDIO_SPPLAY_START 0x0046 // Send data; NetAudio(PC- >AIBO) start #define APPCMD_AUDIO_SPPLAY_STOP 0x0047 // Send data; NetAudio(PC- >AIBO) stop #define APPCMD_AUDIO_SPPLAY_CHMOD 0x0048 // Send data; NetAudio(PC- >AIBO) change data mode // APPCMD_AUDIO_RECV_MODE : nParam1 // APPCMD_AUDIO_SPPLAY_CHMOD : nParam1 #define ATP16K_16BIT_STEREO 0 // 16kHz,16Bit,Stereo #define ATP16K_16BIT_MONO_L 1 // 16kHz,16Bit,Mono Left #define ATP16K_16BIT_MONO_R 2 // 16kHz,16Bit,Mono Right #define ATP16K_16BIT_MONO_MIX 3 // 16kHz,16Bit,Mono Mix #define ATP8K_16BIT_STEREO 4 // 8kHz,16Bit,Stereo . . . >>>Udi's "ABO" toolbox for the 2x0 that uses RCodePlus (stereo data is >available, >>>unless the MatLab wrappers strip it to mono) >>>http://www.benreuven.com/micresponse.htm >> >>(according to the schematic diagram) >>Looks like the ABO interface is only exposing the "left" microphone. The >>MatLab glue will have to be tweeked to get both channels. > >I am sorry that I didn't explain my problem clearly. My robot is ERS-7 and I >am sure if I can use the Rcode in this type of Robots. And Now I only use the >sample "SoundRec" provided by sony company to record outer sound. It seemed >that I could only get one signal. Can I control some variables in OPENR to get >both signals from the two microphones. And even more, can I get four signals >from the two pairs of microphones simultaneously? |
| V 1317 What is latency of Remote Framwork tommytao [2004/08/22-15:20] |
| | Response : Edit : Delete |
|
| In
the FAQ , it said that Remote Framwork is not suitable for low latency | application. | | So, i wanna know what is the definition of low latency application. What is | the average and maximum delay time of Remote Framework? | | Thanks. | |
| +- 1342 RE:What is latency of Remote Framwork openrsupport@(Administrator) [2004/09/03-09:14] |
| | Response : Edit : Delete |
|
| Hi, | | With AIBO Remote Framework, you can send camera data, | microphone audio(WAV) data, and sensor data to your PC. | So AIBO and PC should be connected at least as fast as | 10Mbps in your local network. | | 1) This is a good connection example. | AIBO--(Wireless LAN 10M)---AP===(LAN 100/10M)===Windows PC | | 2) This is not a good connection example. | AIBO--(Wireless LAN 10M)---AP===Modem===(Internet 9600bps)=== | ===Modem===Windows PC(RFW server) *This is low latency application. | | 3) But you can control AIBO via internet by using this sample. | AIBO--(Wireless LAN 10M)---AP===(LAN 100/10M)=== ===Windows PC(RFW, HTTP | server)===Modem===(Internet 9600bps)=== ===Modem===Windows PC(WEB Browser) | | The NetControl and TakePicture sample programs (included in the RFW | distribution) have this function. (You need to setup an HTTP server on your PC) | | 4) The following example is an example of low latency. | All data sent and received by AIBO and PC(RFW) flows in the same | wireless LAN(the same electric wave) at 10Mbps. | AIBO----+ | AIBO----| | AIBO----| | AIBO----+--(Wireless LAN 10M)--AP | AIBO----| | AIBO----| | AIBO----| | AIBO----| | PC(RFW)-| | PC(RFW)-| | PC(RFW)-| | PC(RFW)-| | PC(RFW)-| | PC(RFW)-| | PC(RFW)-| | PC(RFW)-+ | | === | About delay time, | it depends on TCP and network stability. | If the network stability is bad, TCP may resend data occasionally. | This will cause some delay. (Max delay time depends on TCP) | | AIBO Remote Framework uses UDP to send camera data and | microphone audio(WAV). If the data cannot reach the PC, | it disappears (UDP doesn't re-send data) until AIBO sends | the next piece of data to your PC. The PC application (RFW) will lose | the data that disappeared. | |
| +- 1343 RE:What is latency of Remote Framwork openrsupport@(Administrator) [2004/09/03-09:16] |
| Response : Edit : Delete |
|
Sorry,
I posted an error. #2 is a HIGH latency application, not a low latency one. |
| V 1334 Error sounds, but no EMON.LOG abw136 [2004/08/30-23:59] |
| | Response : HomePage : Email : Edit : Delete |
|
| I am
running into some problems debugging my AIBO ERS 220. I am without a | wireless card, so all of my debugging is done via file writing to the memory | stick. Unfortunately, after compiling, linking, and copying all necessary | files to the memory stick, I am experiencing some errors. The AIBO plays an | error tone, and shuts down. The problem at hand is that no EMON.LOG file is | written. I was curious to know what types of errors would yeild this outcome. | | Thank you. | |
| +- 1336 RE:Error sounds, but no EMON.LOG AiboPet [2004/08/31-00:43] |
| Response : HomePage : Edit : Delete |
|
I
assume you are using the Open-R SDK. I *STRONGLY* recommend you get a wireless card. Debugging OpenR programs without it is very difficult. (the other alternative is to wire up a serial debug terminal, which is risky and requires soldering to the main PCB) >Unfortunately, after compiling, linking, and copying all necessary >files to the memory stick, I am experiencing some errors. Have you got it to boot before or is this the first attempt? > The AIBO plays an >error tone, and shuts down. The problem at hand is that no EMON.LOG file is >written. I was curious to know what types of errors would yeild this outcome. There are plenty of problem that will occur without an EMON.LOG file, from improper files on the memory stick to corrupted copy protection on the stick (and others) Debug suggestions: What sound does it make? (monster 7 note movie music or 5 note "duh duh duh duh DUHHH" or something else ?) How long from the time you press the start button until the sound starts (in seconds) Have you used that particular PMS stick with other OpenR or RCODE programs before ? |
| V 1321 how can I get the value of pan,tilt1 and tilt2? robottj [2004/08/25-21:38] |
| | Response : Edit : Delete |
|
| Hello,
I want to get the value of the three joints of head,but failed.Can I | just use the function OPENR::GetJointValue ? Thanks for any help! | |
| +- 1327 RE:how can I get the value of pan,tilt1 and tilt2? AiboPet [2004/08/28-00:46] |
| | Response : HomePage : Edit : Delete |
|
| >Hello,
I want to get the value of the three joints of head,but failed.Can I | >just use the function OPENR::GetJointValue ? Thanks for any help! | | That's the way to do it. | You must OpenPrimitive() the proper joint names first. Also decode the | right 'OJointValue' | | Can you please give more detail on what the error is ? sample code ? | |
| +- 1328 RE:how can I get the value of pan,tilt1 and tilt2? AiboPet [2004/08/28-00:49] |
| | Response : HomePage : Edit : Delete |
|
| Additional: | BTW: usually you don't care about the current value other than the startup | case (stretching) -- which can be found in many samples. | | Once initialized (after AIBO stretches) you usually set the destination | position (SetData to move joints) and assume the servos get to the right | position. | | BTW: another case: if you make the joints limp, and then the human can move | the joints and AIBO acts like a multi-dimensional joystick input. | |
| +- 1333 RE:how can I get the value of pan,tilt1 and tilt2? robottj [2004/08/30-22:21] |
| | Response : Edit : Delete |
|
| Firstly,thank
you very much for your help. | I did the same as what you said, but the result seems wrong.For example,I run | the following codes and the value of headTilt1 is -12296. | OJointValue current; | OPENR::GetJointValue(jointID[12],¤t); | headTilt1=degrees(current.value/1000000.0); | As you see ,I need the value of the joints to calculate the camera's position. | Thanks for any help! | >Additional: | >BTW: usually you don't care about the current value other than the startup | >case (stretching) -- which can be found in many samples. | > | >Once initialized (after AIBO stretches) you usually set the destination | >position (SetData to move joints) and assume the servos get to the right | >position. | > | >BTW: another case: if you make the joints limp, and then the human can move | >the joints and AIBO acts like a multi-dimensional joystick input. | |
| +- 1335 RE:how can I get the value of pan,tilt1 and tilt2? AiboPet [2004/08/31-00:37] |
| Response : HomePage : Edit : Delete |
|
Debug
suggestions: Check the return value from GetJointValue Print out the value of 'current.value' before the calculation [-12296 degrees is a very large number of micro-radians] >I need the value of the joints to calculate the camera's position. Don't you know already know the last position you moved the head to ? |
| V 1329 Colour detection AIBOstudent [2004/08/28-22:02] |
| | Response : Email : Edit : Delete |
|
| I am
currently trying to write a program from an ERS-220 Model using the
Open- | R SDK based on the BallTrackingHead example, and I would appreciate if someone | could help me. | | The program involves the AIBO responding in different ways depending on what | colour is detected, namely red or green. | | I have obtained rough YCrCb values for the colours, but distinguishing between | the colours seems to be complicated by a colour rectangle being defined for | each of the 32 Y values- is it necessary to define a colour rectangle for each | of the 32 values or is it sufficient to define a colour rectangle for a subset | of the 32 Y values? | |
| +- 1330 RE:Colour detection AiboPet [2004/08/29-01:14] |
| Response : HomePage : Edit : Delete |
|
>I
have obtained rough YCrCb values for the colours, but distinguishing
between >the colours seems to be complicated by a colour rectangle being defined for >each of the 32 Y values- is it necessary to define a colour rectangle for each >of the 32 values or is it sufficient to define a colour rectangle for a subset >of the 32 Y values? You should set the rectangle for all 32 Y values (ie. it needs all 32 values) To save time and effort, use the same 4 values (CrMin, CrMax, CbMin, CbMax) for all 32 of them. That's what the samples do with a hard coded pinkish color. That remarkably well except in very dark rooms. For more accurate color ranges, I suggest using the values from the TR.CT file in commercial AiboWare for the standard colors (pink, red, blue, yellow-green etc). Check some past posts for how to do this. The most accurate way is to take photos with AIBO's camera using your lighting conditions (with the object reflecting different light intensities - to cover all 32 Y values). That can be a lot of work, and overkill for simple color detection. |
| V 1322 Remote Framework --- GetInputSemantics Fairlawn07 [2004/08/26-11:46] |
| | Response : Email : Edit : Delete |
|
| void CVAIBO::GetInputSemantics( | int *semID, // Semantics data ID | int *pSubInfoArray, // Semantics data sub-information(SubInfo) | int *pEtcInfoArray ) // Semantics data etc-information(EtcInfo) | elements are ComStatus, Level, Pitch, Position) | | Does anyone have documentation about "pEtcInfoArray" parameter ? Specifically | correlation between semIDs and pEtcInfoArray. | |
| +- 1325 RE:Remote Framework --- GetInputSemantics AiboPet [2004/08/28-00:37] |
| Response : HomePage : Edit : Delete |
|
>Does
anyone have documentation about "pEtcInfoArray" parameter ? Specifically >correlation between semIDs and pEtcInfoArray. I believe it is left intentionally vague. I pass in a an array of 4 ints, and ignore the results (IIRC it fills them with zeros - but maybe not all the time) AFAIK all the good stuff is in the 'pSubInfoArray' |
| V 1319 Remote Framework SEMID_BONE Fairlawn07 [2004/08/25-02:05] |
| | Response : Email : Edit : Delete |
|
| In
the Semantics.txt SubInfo[3] for SEMID_BONE is defined as: = (int)
(((int) | state)&0xffff)) + (((int)angle)&0xffff); | | Should it be: = (int) (((int)state)&0xffff)<< 16) + (((int)angle)&0xffff); ? | |
| +- 1324 RE:Remote Framework SEMID_BONE AiboPet [2004/08/28-00:34] |
| Response : HomePage : Edit : Delete |
|
>In
the Semantics.txt SubInfo[3] for SEMID_BONE is defined as: = (int)
(((int) >state)&0xffff)) + (((int)angle)&0xffff); > >Should it be: = (int) (((int)state)&0xffff)<< 16) + (((int)angle)&0xffff); ? Yes and no. Yes the state is in the upper 16 bits (and the angle in the lower 16 bits). Nitpick: However the values defined for the state are 0x00010000 / 0x00020000 So it should be something like: state_mask | (((int)angle)&0xffff) Nit aside, you are correct, the documentation needs to be clarified. |
| V 1294 Questionnaire on Aibo SDE user motivations Marrekie [2004/08/10-21:03] |
| | Response : Edit : Delete |
|
| Dear user of Aibo Software Development Environment, | | In my thesis research on $BA$(Bo-Creation’, I am studying different approaches for | an interactive collaboration between a company and online user groups. | | I developed this questionnaire for all users of Aibo Software Development | Environment, ranging from just readers of the message board to senior and | expert developers of (application) programs. I want to better understand why | and how you are using Aibo Software Development Environment. I am also very | interested your perceived benefits as a user of Aibo Software Development | Environment. | | Similar questionnaires will be posted in message boards of other developer | platforms. | Filling in this questionnaire will contribute to a general understanding of why | and how people are active on developer platforms. | In the end, benefits could also be there for you acting on a developer platform | that takes the elements into account for even more satisfaction and benefits | for you as a user! | | I kindly invite you to fill in the online questionnaire. This is the link: | http://www.surveymonkey.com/s.asp?u=86986585802. | The questionnaire consists of 15 multiple choice questions and you will need | about 5 minutes to complete it. | | Those who are interested in the motivations of fellow platform users can | receive a summary of the results of this questionnaire. You can indicate this | in the last question of the questionnaire. | | With many thanks for your participation, | | Maaike Spoor | Student at the Rijks Universiteit Groningen | Faculty of Technology Management | www.bdk.rug.nl | maaike.spoor@philips.com | |
| +- 1299 RE:Questionnaire on Aibo SDE user motivations AiboPet [2004/08/11-02:59] |
| | Response : HomePage : Edit : Delete |
|
| > I want to better understand why | > and how you are using Aibo Software Development Environment | | Friendly suggestion: | I see many many many questions on "why", and no questions on "how" | | The "AIBO SDE" is a lot of things under one banner. This website, three | radically different approaches (RCODE, RFW, raw OpenR SDK, plus 3rd party | variants), a number of tools choices etc. | | I believe these and other distinctions of "how" the AIBO SDE resources are | being used would be helpful (ie. "what" parts of the available resources are | you talking about...) | |
| +- 1307 RE:Questionnaire on Aibo SDE user motivations marrekie [2004/08/16-21:05] |
| | Response : Edit : Delete |
|
| Firstly,
sorry for the late reply, I have been busy with other surveys. | Secondly, thanks for your suggestion. | | So I have to integrate a question like: | - What different approaches (RCODE, RFW, raw OpenR SDK, plus 3rd party | variants) are you using? | - Why do you make use of that approach? | - What tools are you using? | | Do you have a clue what approaches (and their tools) are the most used and why? | | Thanks, | Maaike | >> I want to better understand why | >> and how you are using Aibo Software Development Environment | > | >Friendly suggestion: | >I see many many many questions on "why", and no questions on "how" | > | >The "AIBO SDE" is a lot of things under one banner. This website, three | >radically different approaches (RCODE, RFW, raw OpenR SDK, plus 3rd party | >variants), a number of tools choices etc. | > | >I believe these and other distinctions of "how" the AIBO SDE resources are | >being used would be helpful (ie. "what" parts of the available resources are | >you talking about...) | |
| +- 1308 RE:Questionnaire on Aibo SDE user motivations AiboPet [2004/08/17-02:33] |
| | Response : HomePage : Edit : Delete |
|
| >Do
you have a clue what approaches (and their tools) are the most used and | why? | ? Isn't that the purpose of the survey ;-? | | ----- | Based on anecdotal evidence (eg: posts on this BBS, other BBSs, email | questions), here are my guesses. | | RCODE (and variants) is certainly the widest use, and easiest to get started. | Many people try it with no programming experience. Many people have tried it, | but few stick with it - since they aren't being paid to do it. That's a much | larger audience who don't usually read/post here (eg: see Aibo-Life.org BBS). | | Raw OpenR SDK has a number of people using it - not sure how many - because it | is a key part of RoboCup dogs. However I don't think many of the RoboCup folks | read/post here. They are the other extreme - hard core programming in C++ and | they *must* use it for their course/competition (paid labor or slave student | labor ;-) | | RFW is brand new, and few are using it (as you can see by the posts). | | This BBS historically has focused on the OpenR SDK programmer (C++). As you | can probably see, the number of people who post here is relatively small. | They have recently opened it up to RCODE (ERS-7) and RFW - that's when they | started using the term "SDE" ("SDE" is a relatively new term as well) | |
| +- 1312 RE:Questionnaire on Aibo SDE user motivations marrekie [2004/08/18-23:07] |
| Response : Edit : Delete |
|
Thank
you aibo pet for your answers. I have to make up my mind whether I have enough time to make another questionnaire.... Maaike |
| V 1311 how can I get the right sensor of the stereo microphones totoduke [2004/08/18-00:37] |
| Response : Email : Edit : Delete |
|
As
we know, AIBO ERS-7 is equipped with stereo microphones located either
side of its head. But in the model information for ERS-7, I can only find one primitive locator for the stereo microphones : PRM:/rs/c1/c2/c3/m1-Mic:M1. How can I get the address of these two stereo microphones using the primitive locator. |
| V 1301 Remote Framework --- VPRSetMode function Fairlawn07 [2004/08/13-00:08] |
| | Response : Email : Edit : Delete |
|
| I
have problem using the VPRSetMode function. I need more info when to
use it | and the meaning of EVPRMODE enumeration parameters. | Thanks. | |
| +- 1310 RE:Remote Framework --- VPRSetMode function openrsupport@(Administrator) [2004/08/17-09:17] |
| Response : Edit : Delete |
|
>I
have problem using the VPRSetMode function. I need more info when to
use it >and the meaning of EVPRMODE enumeration parameters. >Thanks. Hi, Visual Pattern Recognition has various modes. // VPR mode enum EVPRMODE { EVPRMODE0, // normal EVPRMODE1, // Quest beginning mode EVPRMODE2, // Quest middle EVPRMODE3, // Quest finishing mode, SBC EVPRMODE4, // ESCENE EVPRMODE5, // ECARD Tracking EVPRMODE6, // SBC MAX_EVPRMODE }; Please use the normal mode. EVPRMODE0, // normal The other modes are used during self-charging or during special scenarios in AIBO MIND. MODES 1 - 6 are mainly used in AIBO MIND application scenarios, which are optimized for AIBO MIND. They may be deprecated/deleted in the future if we think they are useless for Remote Framework programmers. Thanks, AIBO SDE Support |
| V 1306 Remote Framework ----- SetColor function Fairlawn07 [2004/08/16-09:01] |
| Response : Email : Edit : Delete |
|
While using SetColor it returns -7 (cannot execute). Did anyone tried
this function? Are there limitations when can we use it? Henry |
| V 1259 set image width*height=208*160,but cpu exception happened. aiboKang [2004/07/19-16:37] |
| | Response : Edit : Delete |
|
| Hi, | Knowing from wconsole that the image width*height (corresponding to LAYER_H) | are 208*160, but when I tried to get camera data from aibo(ERS-7),cpu | exception of address error happened.Hower, when setting image width and height | as 176 and 144(they were used on ERS-210), it's ok! This is weird! | Anyone can tell me how this happened? How are camera data Y,U and V arranged ? | To get Y,U,and V of each pixel, how to set image width and height in circular | sentence when programming? | | Thanks a lot for any help. | | kandar | (The article that 2004/07/19-16:37 posted, was modified by Author) | |
| +- 1264 RE:set image width*height=208*160,but cpu exception happened. AiboPet [2004/07/20-01:14] |
| | Response : HomePage : Edit : Delete |
|
| I'm
not exactly sure what you are doing. | | I never set the width/height of the image data. I only get it from the system | (to be sure it is the size I expect), and then decode the data. | | ---- | | Take a look at the "ImageObserver" sample. It does all this and more | (accessing the different layers, conversion to RGB, figuring out the hires | 416*320 image format, ...) | | The only thing that not covered is creating a hires (416*320) color image. | |
| +- 1291 RE:set image width*height=208*160,but cpu exception happened. aiboKang [2004/08/10-00:13] |
| | Response : Edit : Delete |
|
| >I'm
not exactly sure what you are doing. | > | >I never set the width/height of the image data. I only get it from the system | >(to be sure it is the size I expect), and then decode the data. | > | >---- | > | >Take a look at the "ImageObserver" sample. It does all this and more | >(accessing the different layers, conversion to RGB, figuring out the hires | >416*320 image format, ...) | > | >The only thing that not covered is creating a hires (416*320) color image. | | I wanna directly set the width and height of camera image in circular sentence | when programming instead of using yImage.Width() and yImage.Height(). | | According to returned value from the wireless console, if setting | FbkImageLayer ofbkimageLAYER_H, yImage.Height()=160,yImage.Width()=208.if | setting OFbkImageLayer ofbkimageLAYER_M,yImage.Height()=80,yImage.Width()=104. | ( my AIBO is ERS-7) | | When I set width*height=104*80 (corresponding to ofbkimageLAYER_M),AIBO can | run normally.But when setting width*height=208*160(corresponding to | ofbkimageLAYER_H),the cpu exception of address errors happened. The emon.log | file is like this: | | mCOOP exception | [exception info] | time stamp: 0x00000000ed6e09a4 | status:$12: 0x2000ff03, cause:$13: 0x00000014 | badvaddr: $8: 0x0807c711, epc:$14: 0x8086ad44 | [object info] | context: 0x8023e180, state: 0x00000001 | last context: 0x00000000,last thread: 0x00000000 | [initial value info] | sr:$12: 0x0000ff03 | [register dump] | at: r1: 0x808ab210, v0: r2: 0x3c010000, v1: r3: 0x808ab218, a0: r4: 0x808ab24c | a1: r5: 0x0807c709, a2: r6: 0x00000008, a3: r7: 0x3c010000, t0: r8: 0x806a9740 | t1: r9: 0xfffffffc, t2:r10: 0x00000001, t3:r11: 0x00000020, t4:r12: 0x00007f02 | t5:r13: 0x00000002, t6:r14: 0x83f66648, t7:r15: 0x8023e182, s0:r16: 0x00000010 | s1:r17: 0x00000002, s2:r18: 0x808ab244, s3:r19: 0x00000001, s4:r20: 0x808ab210 | s5:r21: 0xfffffffc, s6:r22: 0x80890d88, s7:r23: 0x00000000, t8:r24: 0x83f80000 | t9:r25: 0x8086aba0, gp:r28: 0x80897570, sp:r29: 0x80628448, s8:r30: 0x00000000 | ra:r31: 0x80857d20, hi: : 0x00000000, lo: : 0x00002be0 | | exception code: 5 Address error exception (store) | | [system control co-processor registers] | Index 0x0000000b Random 0x00000013 EntryLo0 0x01c00317 EntryLo1 0x00000001 | Context 0x000403e0 PageMask 0x00000000 Wired 0x0000000c ??? ---------- | BadVAddr 0x0807c711 Count 0x51eeef3b EntryHi 0x80000000 Compare 0xed83f263 | Status 0x2000ff00 Cause 0x1000802c EPC 0x800ab4fc PRId 0x00002751 | Config 0x404664bb LLAddr 0x00088148 ??? ---------- ??? ---------- | XContext 0x000403e0 ??? ---------- ??? ---------- ??? ---------- | ??? ---------- ??? ---------- ECC 0x00000000 CacheErr 0xf0800000 | TagLo 0x00000000 TagHi 0x00000000 ErrorEPC 0xbfc00e14 ??? ---------- | [floating-point general registers] | $f00: 0x00000000 $f01: 0x00000000 $f02: 0xce8adf12 $f03: 0xc033fffc | $f04: 0x7a34b412 $f05: 0x3e5a39dc $f06: 0xe41164ea $f07: 0x3ffffffa | $f08: 0x921a2000 $f09: 0x3fd1a5a0 $f10: 0x12d1e377 $f11: 0x3f9afb68 | $f12: 0xce8adf12 $f13: 0xc033fffc $f14: 0x4c97c609 $f15: 0x3fb2d0a8 | $f16: 0x1676a3b0 $f17: 0x3fc67576 $f18: 0xf37bebd5 $f19: 0x3fdacee9 | $f20: 0x40090100 $f21: 0x01490000 $f22: 0x00001102 $f23: 0x00000000 | $f24: 0x001c4000 $f25: 0x00012a02 $f26: 0x00001008 $f27: 0x00000200 | $f28: 0x002c0430 $f29: 0x80800000 $f30: 0x00288026 $f31: 0x00300001 | [floating-point control registers] | $0:fp_eir: 0x00002750 | $31:fp_csr: 0x01000000 (bit[17-0]: ----------------RN) | [object list] | Name ContextID ThreadID ExecSpcID OID MetaSpace | ------------------------------------------------------------------------------- | systemCore 0x8022a5e0 0x8022a780 0x8022a740 ---------- mCore | systemCore(1) 0x8022b3e0 0x8022b540 0x8022a740 ---------- mCore | mCOOPReflector 0x8022ebc0 0x802303c0 0x8022fbc0 0x8000000b mCore | uniMailer 0x8022ed20 0x80230420 0x8022fc00 0x8000000c mCore | mCOOPFaultHandler 0x8022ee80 0x80230480 0x8022fc40 0x8000000d mCore | mDriveFaultHandler 0x8022efe0 0x802304e0 0x8022fc80 0x8000000e mCore | registryManager 0x8022f140 0x80230540 0x8022fcc0 0x8000000f mCore | addressManager 0x8022f2a0 0x802305a0 0x8022fd00 0x80000010 mCore | kernelModeLib 0x8022f400 0x80230600 0x8022fd40 0x80000011 mCore | mCoreReflector 0x8022f560 0x80230660 0x8022fd80 0x80000012 mCore | idle 0x8022f6c0 0x802306c0 0x8022fdc0 0x80000013 mCore | exceptionHandler 0x8022f820 0x80230720 0x8022fe00 0x80000014 mCore | analyzer 0x8022f980 0x80230780 0x8022fe40 0x80000015 mCore | mDriveReflector 0x80234980 0x802307e0 0x8022fe80 0x80000016 mCore | obletManager 0x80234820 0x80230840 0x8022fec0 0x80000017 mCore | installer 0x802346c0 0x802308a0 0x8022ff00 0x80000018 mCore | driverManager 0x80234560 0x80230900 0x8022ff40 0x80000019 mCore | objectManager 0x80234400 0x80230960 0x8022ff80 0x8000001a mCore | mClassReflector 0x802342a0 0x802309c0 0x8022ffc0 0x8000001b mCore | mSystemReflector 0x80234140 0x80230a20 0x80230000 0x8000001c mCore | mAVFaultHandler 0x80233fe0 0x80230a80 0x80230040 0x8000001d mCore | mAVReflector 0x80233e80 0x80230ae0 0x80230080 0x8000001e mCore | mAVInit 0x80233d20 0x80230b40 0x802300c0 0x8000001f mCore | avManager 0x80233bc0 0x80235340 0x80230100 0x80000020 mCore | eventManager 0x80238180 0x802352e0 0x80230140 0x80000021 mCore | downloader 0x80237c00 0x80235160 0x80230240 0x80000028 mSystem | nonBlockingFileSystemManager 0x80237aa0 0x80235100 0x80230280 0x80000029 mDrive | openrBusManager 0x80237940 0x802350a0 0x802302c0 0x8000002a mDrive | openrBusManager(1) 0x80238bc0 0x8023c2e0 0x802302c0 0x8000002a mDrive | bmnDriver 0x802377e0 0x80235040 0x80230300 0x8000002b mDrive | bmnDriver(1) 0x802392a0 0x80234ce0 0x80230300 0x8000002b mDrive | fbkDriver 0x80237680 0x80234fe0 0x80230340 0x8000002c mDrive | fbkDriver(1) 0x8023dec0 0x8023bda0 0x80230340 0x8000002c mDrive | fbkDriver(2) 0x8023dd60 0x8023bd40 0x80230340 0x8000002c mDrive | cardManager 0x80237520 0x80234f80 0x80230380 0x8000002d mDrive | cardManager(1) 0x8023c680 0x8023bf80 0x80230380 0x8000002d mDrive | memoryStickDriver 0x802373c0 0x80234f20 0x80238b80 0x8000002e mDrive | memoryStickDriver(1) 0x80239400 0x80234d40 0x80238b80 0x8000002e mDrive | fatFileSystem 0x80239980 0x80234ec0 0x80238b40 0x8000002f mCOOP | memoryStickWatcher 0x80239820 0x80234e60 0x80238b00 0x80000030 mDrive | oobjectManager 0x802396c0 0x80234e00 0x80238ac0 0x80000031 mCOOP | aperiosClass 0x80239560 0x80234da0 0x80238a80 0x80000032 mClass | opowerManager 0x80238020 0x80235280 0x80230180 0x80000033 mCOOP | oserviceManager 0x80237ec0 0x80235220 0x802301c0 0x80000034 mCOOP | ovirtualRobot 0x80237d60 0x802351c0 0x80230200 0x80000035 mCOOP | odesignedRobot 0x80239140 0x80234c80 0x80238a40 0x80000036 mCOOP | osystemLogger 0x80238fe0 0x80234c20 0x80238a00 0x80000037 mCOOP | ovirtualRobotComm 0x80238e80 0x80234bc0 0x802389c0 0x80000038 mCOOP | ovirtualRobotAudioComm 0x80238d20 0x8023c340 0x80238980 0x80000039 mCOOP | IPStack 0x8023d180 0x8023c280 0x80238940 0x8000003a mDrive | OrinocoDriver 0x8023d020 0x8023c220 0x80238900 0x8000003b mDrive | OrinocoDriver(1) 0x8023c520 0x8023bf20 0x80238900 0x8000003b mDrive | OrinocoEnabler 0x8023cec0 0x8023c1c0 0x802388c0 0x8000003c mDrive | hookConsoleIO 0x8023cd60 0x8023c160 0x80238880 0x8000003d mDrive | emergencyMonitor 0x8023cc00 0x8023c100 0x80238840 0x8000003e mCOOP | netconf 0x8023caa0 0x8023c0a0 0x80238800 0x8000003f mCOOP | anttcpio 0x8023c940 0x8023c040 0x802387c0 0x80000040 mDrive | hookConsoleIOActivator 0x8023c7e0 0x8023bfe0 0x80238780 0x80000041 mCOOP | powerMonitor 0x8023c3c0 0x8023bec0 0x80238740 0x80000042 mCOOP | vision 0x8023e180 0x8023be60 0x80238700 0x80000043 mCOOP | tcpgw 0x8023e020 0x8023be00 0x802386c0 0x80000044 mCOOP | [stack info] | area: 0x806279e8 - 0x806285e8, sp: 0x80628448 | [stack dump] | 80628408: 8031ed50 00000001 808a9ce8 80000037 P.1.........7... | 80628418: 803347c8 80333830 8031ed50 80420a44 .G3.083.P.1.D.B. | 80628428: 80318808 80000037 00000004 808a9ce0 ..1.7........... | 80628438: 00000100 80628478 00000008 80000022 ....x.b....."... | 80628448: ffe34000 00000000 8081fc40 8081fd64 .@......@...d... | 80628458: 80897570 00000000 00000003 806a9740 pu..........@.j. | 80628468: 00000009 806284d0 806284e0 808ab4cc ......b...b..... | 80628478: 80897570 80857d20 80890d88 80628558 pu.. }......X.b. | 80628488: 8063d6e8 805de840 80897570 8081fe64 ..c.@.].pu..d... | 80628498: 000000a9 80897570 808108ac 80000022 ....pu......"... | 806284a8: 80897570 80897570 80897570 8081076c pu..pu..pu..l... | 806284b8: 80897570 000001c2 807b6ca0 000000d0 pu.......l{..... | 806284c8: 000000a0 00000410 807b6d70 000000d0 ........pm{..... | 806284d8: 000000a0 00000410 807b6e40 000000d0 ........@n{..... | 806284e8: 000000a0 00000410 80890d88 80628558 ............X.b. | 806284f8: 8063d6e8 805de840 696e6974 80100860 ..c.@.].tini`... | 80628508: 805de7e0 80897570 808111c8 8081119c ..].pu.......... | 80628518: 8063d6e8 80628558 803c9ac8 806287f8 ..c.X.b...<...b. | 80628528: 80897570 80628564 80628558 00000001 pu..d.b.X.b..... | 80628538: 80897570 80808be4 80808dd4 80808f5c pu..........\... | 80628548: 696e6974 6a7b8c9d 80897570 6a7b8c9d tini..{jpu....{j | 80628558: 00000000 80000038 00000009 00000004 ....8........... | 80628568: 00000001 00000001 80628ab4 00000000 ..........b..... | 80628578: 00000003 00000000 00000004 00000002 ................ | 80628588: 80000031 802a0198 00000008 802a022c 1.....*.....,.*. | 80628598: 00000004 00000004 00000000 2f44494f ............OID/ | 806285a8: 0000000c 80897570 80808f94 00726567 ....pu......ger. | 806285b8: 802179dc 80226688 00000003 802a0228 .y!..f".....(.*. | 806285c8: 00000106 80000038 00000009 00010001 ....8........... | 806285d8: 803c9ac8 807b6aa0 00000000 00042240 ..<..j{.....@".. | | BTW,in my program the '.ocf' file was like : | object objectname 3072 16384 128 cache tlb user | | So why and how did that happen? | | Thank you for any help. | | Kandar | |
| +- 1292 RE:set image width*height=208*160,but cpu exception happened. AiboPet [2004/08/10-01:05] |
| Response : HomePage : Edit : Delete |
|
>>I'm
not exactly sure what you are doing. ... >I wanna directly set the width and height of camera image in circular sentence >when programming instead of using yImage.Width() and yImage.Height(). Sorry, I don't understand what that means. --- > But when setting width*height=208*160(corresponding to > ofbkimageLAYER_H),the cpu exception of address errors happened. >... > exception code: 5 Address error exception (store) It looks like your buffer is too small, or there is some other problem with your code reading/decoding the data. Please include a code snippet showing how you are allocating and accessing your buffer. |
| V 1289 Cannot run HelloWorld Sample Program sgutkind [2004/08/06-08:10] |
| | Response : Email : Edit : Delete |
|
| I am
on Linux. I have installed the OPEN-R SDK successfully, but there is an | error when I want to make HelloWorld. Any ideas? | |
| +- 1290 RE:Cannot run HelloWorld Sample Program TodboT [2004/08/07-10:29] |
| Response : Edit : Delete |
|
I'm
not using Linux, so I don't know if I can help, but you're going to
have to have to be a little more specific so others may be able to figure out what's wrong. Let's start with what error messages do you get, and what type of environment you're using. |
| V 1097 How to run a real game with my ERS-7 ? aiboKang [2004/05/22-13:53] |
| | Response : Edit : Delete |
|
| Hi, | | It's exitcing for me to play a real game with my ERS-7, but I don't know how | to make my ERS-7 work with gamecontroller correctly. I tried to run the latest | gamecontroller(RoboCup-GameController-2004-r1 release) in the cywin | environment but failed. It seemed to have some bugs in the gamecontroller, | which was loaded from robocup legged robot official website. It seems | impossible that I got a wrong gamecontroller. Then why it cannot work ? | I'm not familiar with 'makefile', should the 'makefile' be modified ? | | anyway, what should i do when playing a real game ? how can i run the | gamecontroller with my ERS-7 correctly? | | Thanks very much for any help. | |
| |- 1103 RE:How to run a real game with my ERS-7 ? openrsupport@(Administrator) [2004/05/25-10:11] |
| | | Response : Edit : Delete |
|
| | Thank
you for your inquiry. This BBS is intended to support OPEN-R SDK | | users; unfortunately, we cannot answer questions regarding the workings | | of the actual RoboCup games. | | | | We recommend that you review the Four-Legged League's website at: | | http://www.openr.org/robocup/index.html | | If the information on that site does not help you, at the bottom of that | | page, you will find a link to a mailing list where you will be able to | | get answers to your questions. | | | | Thank you, | | | | OPEN-R SDK Support | | |
| | +- 1104 RE:How to run a real game with my ERS-7 ? aiboKang [2004/05/25-11:53] |
| | Response : Edit : Delete |
|
| >We
recommend that you review the Four-Legged League's website at: | >http://www.openr.org/robocup/index.html | >If the information on that site does not help you, at the bottom of that | >page, you will find a link to a mailing list where you will be able to | >get answers to your questions. | > | >Thank you, | > | >OPEN-R SDK Support | | anyway, it's a helpful clue for me , thanks again. | |
| +- 1119 RE:How to run a real game with my ERS-7 ? GTleader [2004/05/29-05:58] |
| | Response : Edit : Delete |
|
| The
RoboCupGameController is used in the Sony Four-Legged Robot League in | RoboCup to send referee decisions to the robots. It cannot be used with the | normal Aibo software. It only runs together with the Open-R SDK. | |
| +- 1121 RE:How to run a real game with my ERS-7 ? aiboKang [2004/05/31-18:27] |
| | Response : Edit : Delete |
|
| >The
RoboCupGameController is used in the Sony Four-Legged Robot League in | >RoboCup to send referee decisions to the robots. It cannot be used with the | >normal Aibo software. It only runs together with the Open-R SDK. | | But It doesn't work. I mean I cannot build the RoboCupGameController | (RoboCupGameController-2004-r2 downloaded from Robocup Official website).When | I tried to compile the RoboCupGameController-2004-r2, some errores happened.It | seems that the RobopCupGameController provided by RoboCup Legged Robot League | should not have bugs.So I don't know why I can't build it. Are there something | wrong with my cygwin ? or else ? | BTW, what size is the ball used in RoboCup ? | Best regards. | | Kandar | |
| +- 1156 RE:How to run a real game with my ERS-7 ? GTleader [2004/06/11-01:43] |
| | Response : Edit : Delete |
|
| You
are right. For some strange reasons, two empty directories were missing
in | the archive. It seems that they were lost when the archive was unpacked with | the wrong tool and then re-packed again. | | At least with the Cygwin/Linux "tar" it should work now. | | Thomas | |
| +- 1221 RE:How to run a real game with my ERS-7 ? aiboKang [2004/07/02-02:41] |
| | Response : Edit : Delete |
|
| >You
are right. For some strange reasons, two empty directories were missing
in | >the archive. It seems that they were lost when the archive was unpacked with | >the wrong tool and then re-packed again. | > | >At least with the Cygwin/Linux "tar" it should work now. | > | >Thomas | | I unpacked gamecontroller archive with "tar" in cywgin,but still failed when I | tried to build it. | I unpacked gamecontroller archive like this (in cygwin ) : | ------ $ tar zxvf RoboCup-GameController-2004-r2.tar.gz | then I tried to build it: | --------$ cd RoboCup-GameController-2004-r2/robocupgamecontroller/rp/host | --------$ make | and then error happened like this: | | $ make | for c in ../../RoboCupGameController ../../RoboCupIPConfigurator; do \ | make -C $c/host INSTALLDIR=/gc2004/robocup-gamecontroller-2004- | r2/robocu | pgamecontroller/rp/host/MS all ; \ | done | make[1]: Entering directory `/gc2004/robocup-gamecontroller-2004- | r2/robocupgamec | ontroller/RoboCupGameController/host' | /usr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-stubgen2 ../stub.cfg | Object = RoboCupGameController | numOfSubject = 8 | numOfObserver = 1 | Service name = "RoboCupGameController.PlayerA1.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerA2.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerA3.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerA4.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerB1.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerB2.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerB3.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.PlayerB4.RoboCupGameControlData.S" | Control Function = null | Ready Function = null | Service name = "RoboCupGameController.Dummy.DummyData.O" | Connect Function = null | Notify Function = null | patch < stub.patch | patching file RoboCupGameControllerStub.cc | g++ -Wno-deprecated -I. -I.. -I../../include `pkg-config gtk+-2.0 --cflags` | `/u | sr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-config --cflags` -o | RoboCupGameContro | llerStub.o -c RoboCupGameControllerStub.cc | g++ -Wno-deprecated -I. -I.. -I../../include `pkg-config gtk+-2.0 --cflags` | `/u | sr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-config --cflags` -o | RoboCupGameContro | ller.o -c ../RoboCupGameController.cc | g++ -Wno-deprecated -I. -I.. -I../../include `pkg-config gtk+-2.0 --cflags` | `/u | sr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-config --cflags` -o GUI.o - | c ../GUI.c | c | g++ -Wno-deprecated -I. -I.. -I../../include `pkg-config gtk+-2.0 --cflags` | `/u | sr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-config --cflags` -o Penalty.o - | c ../P | enalty.cc | g++ -Wno-deprecated -I. -I.. -I../../include `pkg-config gtk+-2.0 --cflags` | `/u | sr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-config --cflags` - | o ../../libTeamInfo | /src/TeamInfo.o -c ../../libTeamInfo/src/TeamInfo.cc | g++ -o roboCupGameController.bin RoboCupGameControllerStub.o | RoboCupGameControl | ler.o GUI.o Penalty.o ../../libTeamInfo/src/TeamInfo.o `pkg-config gtk+-2.0 -- | li | bs` `/usr/local/OPEN_R_SDK/RP_OPEN_R/bin/rp-openr-config --libs` | /usr/lib/gcc-lib/i686-pc-cygwin/3.3.1/../../../../i686-pc-cygwin/bin/ld: cannot | find -lintl | collect2: ld returned 1 exit status | make[1]: *** [roboCupGameController.bin] Error 1 | make[1]: Leaving directory `/gc2004/robocup-gamecontroller-2004- | r2/robocupgameco | ntroller/RoboCupGameController/host' | make[1]: Entering directory `/gc2004/robocup-gamecontroller-2004- | r2/robocupgamec | ontroller/RoboCupIPConfigurator/host' | make -C .. all | make[2]: Entering directory `/gc2004/robocup-gamecontroller-2004- | r2/robocupgamec | ontroller/RoboCupIPConfigurator' | g++ -O2 -I. -I../include `pkg-config gtk+-2.0 --cflags` -o | RoboCupIPConfigurator | .o -c RoboCupIPConfigurator.cc | g++ -O2 -I. -I../include `pkg-config gtk+-2.0 --cflags` -o Gui.o -c Gui.cc | g++ -o RoboCupIPConfigurator RoboCupIPConfigurator.o | Gui.o ../libTeamInfo/src/Te | amInfo.o `pkg-config gtk+-2.0 --libs` | /usr/lib/gcc-lib/i686-pc-cygwin/3.3.1/../../../../i686-pc-cygwin/bin/ld: cannot | find -lintl | collect2: ld returned 1 exit status | make[2]: *** [RoboCupIPConfigurator] Error 1 | make[2]: Leaving directory `/gc2004/robocup-gamecontroller-2004- | r2/robocupgameco | ntroller/RoboCupIPConfigurator' | make[1]: *** [all] Error 2 | make[1]: Leaving directory `/gc2004/robocup-gamecontroller-2004- | r2/robocupgameco | ntroller/RoboCupIPConfigurator/host' | make: *** [all] Error 2 | | I don't understand why and how to correct it. | Expecting your help !! | | kandar | |
| +- 1255 RE:How to run a real game with my ERS-7 ? GTleader [2004/07/16-05:11] |
| | Response : Edit : Delete |
|
| Maybe
something is missing in your cygwin installation. The linker reports | that it cannot find -lintl. This sounds as if /lib/libintl.a | or /lib/libintl.la is missing. | | Is GTK 2 installed? | | Thomas | |
| +- 1258 RE:How to run a real game with my ERS-7 ? aiboKang [2004/07/17-22:32] |
| | Response : Edit : Delete |
|
| >Maybe
something is missing in your cygwin installation. The linker reports | >that it cannot find -lintl. This sounds as if /lib/libintl.a | >or /lib/libintl.la is missing. | > | >Is GTK 2 installed? | > | >Thomas | | yes, I did. | I installed gtk2 from http://web.sfc.keio.ac.jp/~s01397ms/cygwin/ . | I searched my computer for libintl.a, but only found libintl.awk(in | cygwin/usr/share/awk) instead of libintl.a or libintl.la. It seemed as if no | libintl.a on my computer.So if libintl.a or libintl.la should be include in | cygwin, what else should be installed? where can I get it? | thanks. | | kandar | (The article that 2004/07/17-20:58 posted, was modified by Author) | |
| +- 1283 RE:How to run a real game with my ERS-7 ? GTleader [2004/08/01-06:45] |
| | Response : Edit : Delete |
|
| As
far as I can see, libintl is part of the gettext package in cygwin,
which | can easily be revealed by typing "cygwin libintl" into Google. | | Thomas | |
| +- 1288 RE:How to run a real game with my ERS-7 ? aiboKang [2004/08/05-17:53] |
| Response : Edit : Delete |
|
>As
far as I can see, libintl is part of the gettext package in cygwin,
which >can easily be revealed by typing "cygwin libintl" into Google. > >Thomas Solved! At first I installed libintl with cygwin setup,but the problem still existed. Later I directly copy libintl.a to directory /lib/ , and it's done. Thank you very very much,Thomas. Best regards, kandar |
| V 1284 usb memory reader writer navacron [2004/08/03-04:11] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I want to programme an aibo. But i cant figure out what is required to write to | aibo memory stick. Can "Sony Memory Stick$B%g(B USB Reader/Writer MSAC-US300" | product read and write to the aibo memory stick? Can other products like | "SanDisk Memory Stick/Memory Stick Pro USB 2.0 Card Reader" write to memory | stick . Are there any older models of sony usb writer that can work? I would | appreciate what other members are using to write to aibo memory sticks. | | Thanks | Naveed | |
| +- 1286 RE:usb memory reader writer AiboPet [2004/08/04-00:45] |
| | Response : HomePage : Edit : Delete |
|
| >Hi, | > | >I want to programme an aibo. But i cant figure out what is required to write | to | >aibo memory stick. | | Any *Sony* brand stick reader/writer will do. Also Sony cameras and PDAs | (CLIE) can work when connected to your PC. | I strongly recommend you *never* put a PINK memory stick in any reader that is | not made by Sony. | | > Can "Sony Memory Stick$B%g(B USB Reader/Writer MSAC-US300" | The "MSAC-US30" is great. I use it all the time. Very fast when connected with | USB 2. Highly recommended. | | >Can other products like | >"SanDisk Memory Stick/Memory Stick Pro USB 2.0 Card Reader" write to memory | >stick | Stay away from non-Sony brand devices. They may work, but they may corrupt the | stick ! If so there is no way you can fix it. Stick with Sony brand. | | > Are there any older models of sony usb writer that can work? I would | >appreciate what other members are using to write to aibo memory sticks. | The older Sony models (eg: MSAC-US1A, the memory stick mouse, etc) will also | work. The MSAC-US30 is still the best. | |
| +- 1287 RE:usb memory reader writer navacron [2004/08/04-23:24] |
| Response : Email : Edit : Delete |
|
>>Hi, >> >>I want to programme an aibo. But i cant figure out what is required to write >to >>aibo memory stick. > >Any *Sony* brand stick reader/writer will do. Also Sony cameras and PDAs >(CLIE) can work when connected to your PC. >I strongly recommend you *never* put a PINK memory stick in any reader that is >not made by Sony. > >> Can "Sony Memory Stick$B%g(B USB Reader/Writer MSAC-US300" >The "MSAC-US30" is great. I use it all the time. Very fast when connected with >USB 2. Highly recommended. > >>Can other products like >>"SanDisk Memory Stick/Memory Stick Pro USB 2.0 Card Reader" write to memory >>stick >Stay away from non-Sony brand devices. They may work, but they may corrupt the >stick ! If so there is no way you can fix it. Stick with Sony brand. > >> Are there any older models of sony usb writer that can work? I would >>appreciate what other members are using to write to aibo memory sticks. >The older Sony models (eg: MSAC-US1A, the memory stick mouse, etc) will also >work. The MSAC-US30 is still the best. Thankyou! You have been a great help!! I am going to order msac-us30. |
| V 1253 ad-hoc network djtommyc [2004/07/15-02:49] |
| | Response : Email : Edit : Delete |
|
| Can
the Aibo be made to work on an ad-hoc wireless connection with another
computer? | |
| +- 1254 RE:ad-hoc network djtommyc [2004/07/15-04:03] |
| | Response : Email : Edit : Delete |
|
| >Can
the Aibo be made to work on an ad-hoc wireless connection with another | computer? | | nevermind, it is | |
| +- 1280 RE:ad-hoc network djtommyc [2004/07/27-01:28] |
| | Response : Email : Edit : Delete |
|
| >>Can
the Aibo be made to work on an ad-hoc wireless connection with another | >computer? | > | >nevermind, it is | | well I had an ad-hoc connection going with the Aibo for a day. Now, I can't get | it to work anymore. All the settings have remained the same, best I can figure, | I must be getting interferance. Anyone have any ideas? | |
| +- 1282 Wireless Access Point usage openrsupport@(Administrator) [2004/07/28-09:42] |
| Response : Edit : Delete |
|
Many
new notebook PCs have wireless LAN functionality built-in. You can
connect AIBO and your PC via Ad Hoc mode (IBSS peer to peer) without a Wireless Access Point. However, we recommend using a Wireless Access Point in the following cases: 1. To make setting up your wireless LAN easy. 2. There is another wireless LAN (.11b or .11g) near you. In this case, you can see the other Access Point(s) in the list of available networks in the Wireless Network Connection Properties. When using Ad Hoc mode, the wireless LAN channel is fixed, which can make it jam easily. 3. To connect your PC to the Internet. It is hard to set up dual routing for PtoP and an Internet connection. 4. To use other wireless LAN devices in addition to one AIBO and one PC. If the connection occasionally becomes unavailable, it may be caused by 2. AIBO SDE Support |
| V 1274 Can someone give a simple explain or example of IsReady()? wangy22 [2004/07/22-09:52] |
| | Response : Edit : Delete |
|
| Hi. | | I want to control the execution order of DoStart() functions in all my OPEN-R | objects. I check the reference guide and found IsReady() seems has something to | do with this. But I am still unsure how to use it. | | What are the conditions for IsReady() to return different values? | | Thanks! | | Yu | |
| +- 1276 RE:Can someone give a simple explain or example of IsReady()? AiboPet [2004/07/24-01:03] |
| | Response : HomePage : Edit : Delete |
|
| ?
Can someone give a simple explaination of IsReady ? | | No, not simple ;-> | | ---- | IsReady is a simple concept, it is true when an observing object is ready to | receive a data notification, false otherwise. | In general you don't need to test it yourself, because the object | notifications callback handle that stuff for you (MyClass::???Ready) | | The tricky problem is the startup case. | Look at any of the examples, eg: MovingLegs7 | Usually this is in the connection to "OVirtualRobotComm.Effector", which | controls the robot's legs/head etc [ie. MovingLegs7 sends control messages to | the Effector to move the legs] | | When the MovingLegs7 object starts up, the 'DoStart' function is called. | If the Effector is ready, the FIRST movement command is sent to the Effector, | and begins the startup sequence (usually stretching the legs). Additional | movement commands occur later after the FIRST command is completed (see | the 'Ready' callback function) | | HOWEVER, if the Effector is *NOT* ready yet, then no commands are sent. | Eventually the 'Ready' callback will be called, and the FIRST movement command | can be sent. | | To be general, your code should handle both cases. The startup order of | objects can not normally be guaranteed. | |
| +- 1278 RE:Can someone give a simple explain or example of IsReady() wangy22 [2004/07/25-02:09] |
| | Response : Edit : Delete |
|
| >?
Can someone give a simple explaination of IsReady ? | > | >No, not simple ;-> | > | >---- | >IsReady is a simple concept, it is true when an observing object is ready to | >receive a data notification, false otherwise. | >In general you don't need to test it yourself, because the object | >notifications callback handle that stuff for you (MyClass::???Ready) | > | >The tricky problem is the startup case. | >Look at any of the examples, eg: MovingLegs7 | >Usually this is in the connection to "OVirtualRobotComm.Effector", which | >controls the robot's legs/head etc [ie. MovingLegs7 sends control messages to | >the Effector to move the legs] | > | >When the MovingLegs7 object starts up, the 'DoStart' function is called. | >If the Effector is ready, the FIRST movement command is sent to the Effector, | >and begins the startup sequence (usually stretching the legs). Additional | >movement commands occur later after the FIRST command is completed (see | >the 'Ready' callback function) | > | >HOWEVER, if the Effector is *NOT* ready yet, then no commands are sent. | >Eventually the 'Ready' callback will be called, and the FIRST movement command | >can be sent. | > | >To be general, your code should handle both cases. The startup order of | >objects can not normally be guaranteed. | | Thanks, that helps me a lot. | | Can I control the DoStart() startup order for all objects by using IsReady()? | |
| +- 1279 RE:Can someone give a simple explain or example of IsReady() AiboPet [2004/07/27-00:39] |
| | Response : HomePage : Edit : Delete |
|
| >
Can I control the DoStart() startup order for all objects by using
IsReady()? | | Well you aren't controlling the startup order of the objects. | The 'IsReady' logic is meant to handle any possible startup order, and the | multi-tasking nature of the operating system. | The technique can be used to fix other possible startup conflicts [but can be | hard to test] | | The order can be random, and what happens before an object is ready to receive | notifications depends on how the object is implemented. The order of the | OBJECT.CFG file usually decides the order, but it isn't always that simple. | | Other alternatives: | [ie. the fewer calls to IsReady, the better IMHO] | If you are building more complicated objects communicating to eachother, | another solution is to not send notifications until everything is up and | running (ie. all objects have returned from DoStart). | Then you can have one object start the real behavior. | | Finally there is an undocumented? feature of the .OCF file where you can set | the priority for an object. The default is 128. Objects with larger numbers | are loaded first. I would only play with that as a last resort if you have | race condition problems. | |
| +- 1281 RE:Can someone give a simple explain or example of IsReady() wangy22 [2004/07/27-03:53] |
| Response : Edit : Delete |
|
>>
Can I control the DoStart() startup order for all objects by using
IsReady()? > >Well you aren't controlling the startup order of the objects. >The 'IsReady' logic is meant to handle any possible startup order, and the >multi-tasking nature of the operating system. >The technique can be used to fix other possible startup conflicts [but can be >hard to test] > >The order can be random, and what happens before an object is ready to receive >notifications depends on how the object is implemented. The order of the >OBJECT.CFG file usually decides the order, but it isn't always that simple. > >Other alternatives: >[ie. the fewer calls to IsReady, the better IMHO] >If you are building more complicated objects communicating to eachother, >another solution is to not send notifications until everything is up and >running (ie. all objects have returned from DoStart). >Then you can have one object start the real behavior. > >Finally there is an undocumented? feature of the .OCF file where you can set >the priority for an object. The default is 128. Objects with larger numbers >are loaded first. I would only play with that as a last resort if you have >race condition problems. Thank you for the help! |
| V 1275 Can Program Memory stick read write fr PDA MS slot? aiboselva [2004/07/22-15:41] |
| | Response : HomePage : Email : Edit : Delete |
|
| hi
Is it possible to save r script on aibo program memory stick by putting
it | in sony pda memory stick slot?I will be developing a palm program which will | create r script so i was wondering whether aibo program MS will fit into pda MS | slot and whether the script can be copied directly to programMS? | |
| +- 1277 RE:Can Program Memory stick read write fr PDA MS slot? AiboPet [2004/07/24-01:06] |
| Response : HomePage : Edit : Delete |
|
>hi
Is it possible to save r script on aibo program memory stick by putting
it >in sony pda memory stick slot?I will be developing a palm program which will >create r script so i was wondering whether aibo program MS will fit into pda MS >slot and whether the script can be copied directly to programMS? Yes (Sony Brand CLIE PDAs or cameras can be used) One way is to run the "MS Export" (or equivalent) program on the PDA when hooked up to the PC's USB cable. The memory stick will appear on the PC as another drive letter. Use it as a standard stick reader. For writing your own CLIE programs, you can access the memory stick files and data using the VFS PalmSource APIs. The volume type of a memory stick is 'mstk' |
| V 1271 Segmentation Fault on ASSERT_READY_TO_ALL_OBSERVER wangy22 [2004/07/22-05:30] |
| | Response : Edit : Delete |
|
| Hi, | | I've just implemented a simple motion manager similar to MotionAgent in MoNet | example. | | However, I got a segmentation fault just after the AIBO booting and found the | problem here: | | wangy22@linux gMotion $ emonLogParser emon.log gMotion.nosnap.elf | context: 802620e0 | object: gMotion | badvaddr: 00000018 | epc: 22704b68 | ra: 22701160 | target address: 22704b68 (epc) | gp: 2277ef60 | _gp: 0047ef60 | static addr: 00404b68 | symbol: 00404b40 T OObserver::SendReadyAll(int) | | I highly suspect this SendReadyAll() function is called in | ASSERT_READY_TO_ALL_OBSERVER in DoStart() stage. | | What are the possible reasons causing this error? | Can someone help me figure it out? | | Thanks a lot! | |
| |- 1272 Here are my CONNECT.CFG and stub.cfg wangy22 [2004/07/22-05:39] |
| | Response : Edit : Delete |
|
| #my CONNECT.CFG: | #---------------------- | #gServer --> gHead | gServer.gHeadSend.gHeadCommand.S gHead.gHeadRecv.gHeadCommand.O | | #gHead --> OVirtualRobotComm | gHead.Control.OCommandVectorData.S OVirtualRobotComm.Effector.OCommandVectorData.O | | #gMotion --> gJoints | gMotion.gJointsSend.gJointsCommand.S gJoints.gJointsRecv.gJointsCommand.O | | #gJoints --> OVirtualRobotComm | gJoints.Control.OCommandVectorData.S OVirtualRobotComm.Effector.OCommandVectorData.O | | #gServer --> gMotion | gServer.gMotionSend.gMotionCommand.S gMotion.gMotionRecv.gMotionCommand.O | | #stub.cfg | #------------------------- | ObjectName : gMotion | NumOfOSubject : 1 | NumOfOObserver : 1 | Service : "gMotion.gJointsSend.gJointsCommand.S", null, Ready() | Service : "gMotion.gServerRecv.gMotionCommand.O", null, gServerNotify() | |
| +- 1273 PROBLEM SOLVED wangy22 [2004/07/22-08:57] |
| Response : Edit : Delete |
|
Problem
was sovled by moving all the initialization statements after ENABLE_ALL_SUBJECT and ASSERT_READY_TO_ALL_OBSERVER. Those statements initializes TCP/IP network. |
| V 1251 What happens if SetMotorPower(opowerON) is called twice? wangy22 [2004/07/13-05:54] |
| | Response : Edit : Delete |
|
| I
notice that in the OPEN-R SDK Samples, SetMotorPower(opowerON) is not
called | if there is another object calling the function. I am wondering what happens if | SetMotorPower(opowerON) is called again? | | Thanks! | |
| +- 1270 RE:What happens if SetMotorPower(opowerON) is called twice? openrsupport@(Administrator) [2004/07/22-02:09] |
| Response : Edit : Delete |
|
Hi, If SetMotorPower(opowerON) is called twice, the second invocation is a no-op. It's harmless. Thanks, AIBO SDE Support >I notice that in the OPEN-R SDK Samples, SetMotorPower(opowerON) is not called >if there is another object calling the function. I am wondering what happens if >SetMotorPower(opowerON) is called again? > >Thanks! |
| V 1256 Debian GNU/Linux Compile TCPGateway Error robottj [2004/07/16-19:48] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
use Deiban GNU/Linux as my develop flat.These is my system information: | | uname -a | Linux tanfeng.localdomain 2.4.26-1-686-smp #1 SMP Sat May 1 19:17:11 EST 2004 | i686 GNU/Linux | | gcc --version | gcc (GCC) 3.3.4 (Debian 1:3.3.4-2) | | I follow the instruction in README file for TCPGateway for RoboCup2003 - release 1 | | "INSTALL | | ------- | | This package includes only two .o files. You have to re-compile tcpGateway | | executable as the following. | | % cd /usr/local/OPEN_R_SDK | | % rm RP_OPEN_R/lib/linux-i386/TCPGateway/tcpGateway | | % tar xzvf somewhere/TCPGateway-patch-for-RoboCup2003-r1.tar.gz | | % setup-rp-openr | | " | | When I run script setup-rp-openr | it gives these errors: | | tanfeng:/usr/local/OPEN_R_SDK/RP_OPEN_R/bin# ./setup-rp-openr | | make: Nothing to be done for `all'. | make: Nothing to be done for `all'. | make: Nothing to be done for `all'. | g++ -o tcpGateway TCPGatewayStub.o TCPGateway.o ../libObjectComm.a | ../libOPENR.a -lpthread | TCPGatewayStub.o(.gnu.linkonce.t._ZNSt13_Format_cacheIcE6_S_getERSt8ios_base+0x75): | In function `std::_Format_cache<char>::_S_get(std::ios_base&)': | : undefined reference to `std::_Format_cache<char>::_Format_cache[in-charge]()' | TCPGatewayStub.o(.gnu.linkonce.t._ZNSt13_Format_cacheIcE11_S_callbackENSt8ios_base5eventERS1_i+0xfe): | In function `std::_Format_cache<char>::_S_callback(std::ios_base::event, | std::ios_base&, int)': | : undefined reference to `std::_Format_cache<char>::_Format_cache[in-charge]()' | collect2: ld returned 1 exit status | make: *** [tcpGateway] Error 1 | | I think it may be because of the library pthread.Can Anyone who use Debian tell | me how to resolve this problem? Thanks | |
| +- 1269 RE:Debian GNU/Linux Compile TCPGateway Error openrsupport@(Administrator) [2004/07/22-02:08] |
| Response : Edit : Delete |
|
Thank
you for your inquiry. This BBS is intended to support OPEN-R SDK users; unfortunately, we cannot answer questions regarding the workings of the actual RoboCup games. We recommend that you review the Four-Legged League's website at: http://www.openr.org/robocup/index.html If the information on that site does not help you, at the bottom of that page, you will find a link to a mailing list where you will be able to get answers to your questions. Thank you, OPEN-R SDK Support |
| V 1267 Motion Editor rosaluna [2004/07/22-00:28] |
| | Response : Edit : Delete |
|
| How
can I use the motions made by the AIBO Motion Editor with OPEN-R? | |
| +- 1268 RE:Motion Editor AiboPet [2004/07/22-01:32] |
| Response : HomePage : Edit : Delete |
|
>How
can I use the motions made by the AIBO Motion Editor with OPEN-R? The new MEdit motion editor (or the free Skitter motion editor or other editors for older models) all use the same .MTN motion format (.MTN files are usually combined together into one big .ODA file) .MTN playback is built-in to RCODE. .MTN playback is not built-in to the OpenR SDK, but it is provided as part of the samples (include the MoNet module and glue it together). See the BallTrackingHead7 sample for one way to do this. Part of message 1098 https://openr.aibo.com/cgi-bin/openr/e_regi/im_trbbs.cgi? uid=general&df=bbs.dat&prm=TAN&pg=-1&no=1098#1098 |
| V 1265 DoApproach function in the Remote Framework Fairlawn07 [2004/07/21-02:38] |
| | Response : Email : Edit : Delete |
|
| Execution
of the DoApproach function doesn't cause AIBO to move at all??? | |
| +- 1266 RE:DoApproach function in the Remote Framework TodboT [2004/07/21-07:28] |
| Response : Edit : Delete |
|
>Execution
of the DoApproach function doesn't cause AIBO to move at all??? I haven't used this command, but you can take a look at the RemoteTest sample. For aibo to walk you need to specify a target and aibo needs to see the target before it'll actually move. In the RemoteTest app you can have aibo Search & Track a target (Kick/Header.. button). Then when the target is being tracked you can have aibo walk forward with the TARGET specified (Walk button). Now aibo should approach the target. |
| V 1261 Using wireless lan cards other than ERA-201D on Aibo ERS210? wangy22 [2004/07/19-01:09] |
| | Response : Edit : Delete |
|
| Since
the ERA-201D AIBO Wireless LAN Card is discontinued by SONY, I am | wondering if it is possible to use wireless card other than ERA-201D in AIBO | ERS-210? | | I tried to insert ERA-201D card into my laptop and it works perfect under Linux. | The driver modules it uses are orinoco_cs and hermes (kernel 2.6.5). Luckily, I | found the word "orinoco" in the wireless console too, when AIBO is booting. | | So I am thinking the possiblity of using a wireless card with the same chipset | (orinoco 802.11b) in AIBO. | | Anyone? | |
| +- 1263 RE:Using wireless lan cards other than ERA-201D on Aibo ERS210? AiboPet [2004/07/19-03:26] |
| Response : HomePage : Edit : Delete |
|
>Since
the ERA-201D AIBO Wireless LAN Card is discontinued by SONY, I am >wondering if it is possible to use wireless card other than ERA-201D in AIBO >ERS-210?... The chipset and form factor must be the same. The only known replacement/alternative is the Sony "PCWA-C150S" (also discontinued). The current model Sony PCMCIA card doesn't work (they changed the chipset to support 802.11g) http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi? ubb=get_topic;f=10;t=000020 If you can find someone with old stock of the "C150S" card, then consider yourself lucky. Also the old "AiboEyes" package included the official "ERA- 201D" card. Last resort: The official cards are still available in Japan (expect $$$ broker/import fees) http://www.jp.aibo.com/products/acc_ers_200.html |
| V 1260 Still installation problem, on Linux A_Banknote [2004/07/18-21:08] |
| | Response : Edit : Delete |
|
| I'm on a Linux. | And I cannot compile any samples. | All got a mistake which said there is no mipsel-linux-g++ in the | /usr/local/OPEN_R_SDK/bin | I think the mipsel-linux-g++ is only the part of cygwin on a Windows. | I do have the suit of gcc-3.3.3-r6. | How can I complie any samples? | Is the mipsel-linux-g++ needed? | Do I have to run that script which makes all the suit of gcc-3.3.2 ? | |
| +- 1262 RE:Still installation problem, on Linux wangy22 [2004/07/19-01:10] |
| Response : Edit : Delete |
|
>I'm on a Linux. >And I cannot compile any samples. >All got a mistake which said there is no mipsel-linux-g++ in the >/usr/local/OPEN_R_SDK/bin >I think the mipsel-linux-g++ is only the part of cygwin on a Windows. >I do have the suit of gcc-3.3.3-r6. >How can I complie any samples? >Is the mipsel-linux-g++ needed? >Do I have to run that script which makes all the suit of gcc-3.3.2 ? Check this out: http://www-2.cs.cmu.edu/~tekkotsu/SDKInstall.html It might be helpful. |
| V 1252 about the memory totoduke [2004/07/13-05:54] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I found that I could not allocate a space for a 2-D matrix, which has a size | of [2][9025]. I am not sure if it is forbidden to open a larger matrix or the | memory is not enough to allocate such a space. Could anyone tell me what's | happening. Thanks a lot | |
| +- 1257 RE:about the memory wangy22 [2004/07/17-05:08] |
| Response : Edit : Delete |
|
>Hi, > >I found that I could not allocate a space for a 2-D matrix, which has a size >of [2][9025]. I am not sure if it is forbidden to open a larger matrix or the >memory is not enough to allocate such a space. Could anyone tell me what's >happening. Thanks a lot What is the data type of your 2-D matrix? |
| V 1250 Administrator DELETE ***** [2004/07/14-09:31] |
| Response : Edit : Delete |
|
(The
article that 2004/07/13-05:53 posted, was modified by Administrator) |
| V 1245 Applications for Remote Framework Gilles [2004/07/10-06:14] |
| | Response : Edit : Delete |
|
| For
use of Remote Framework I read that Visual C++ 6.0 (or Visual C++ .NET | 2003) is necessary. | Is there a free version of these softwares or I must buy them ? | |
| +- 1247 RE:Applications for Remote Framework AiboPet [2004/07/11-01:10] |
| | Response : HomePage : Edit : Delete |
|
| >For
use of Remote Framework I read that Visual C++ 6.0 (or Visual C++ .NET | >2003) is necessary. | >Is there a free version of these softwares or I must buy them ? | | No, those are not free. | The full versions are needed to build the samples. | That's recommended if you want to make full use of RFW. | | ----- | If you are creating your own command line programs from scratch (using C++, | not using the samples or MFC) then you can use the free Microsoft compiler. | http://msdn.microsoft.com/visualc/vctoolkit2003 | [NOTE: Will not build the RFW samples] | | ------ | Finally, if you only want to run the programs, you don't need any compiler. | |
| +- 1249 Author DELETE ***** [2004/07/13-01:14] |
| Response : Edit : Delete |
|
(The
article that 2004/07/11-16:18 posted, was modified by Author) |
| V 1246 VB interface for R-Code??? dharry [2004/07/10-20:05] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hi World, | | I have just bought a 7 and would like to start doing some programming. Problem | is is that I don't know C.... | | VB or VB.NET are no problem....has anyone managed to do any conversions to VB? | | Or know anywhere that may be of help... | | Regards, | Dean | |
| +- 1248 RE:VB interface for R-Code??? AiboPet [2004/07/11-01:15] |
| Response : HomePage : Edit : Delete |
|
> >I have just bought a 7 and would like to start doing some programming. Problem >is is that I don't know C.... I suggest you start with YART. http://www.aibohack.com/ers7/yart7.htm It helps you get into R-CODE programming easily (using a GUI tool). Then you can decide if you want to manually edit the R-CODE.R file, or do other approaches. The R-CODE syntax has a flavor like BASIC (but is very different in many ways). --- BTW: For more elaborate scenarios, if you have a lot of code in VB you can run it on your PC, then send RCODE command strings to the AIBO via WiFi. That's how many simple programs like AiboRemote work. The RCODE command protocol is very simple (just like telnet). |
| V 1082 DNSLookUp console echo problem angel [2004/05/13-04:23] |
| | Response : HomePage : Email : Edit : Delete |
|
| While
around with the DSNLookUp demo of the samples, I noticed that every time | one types an address in the console AIBO responds with an echo. For example if | you type 123.123.123.123 at the console prompt this is what you would see: | | DNSLookUp> 112233..112233..112233..112233 | | Now I have look at the code heads and tails and I cant see anywhere an echo | could be generating unless this is somehow a much lower level bug. Anyone have | seen this before? | |
| |- 1083 More information on DNSLookUp angel [2004/05/13-08:24] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
narrow down the problem to the DNSLookUp::Prompt() method, specifically
to the | | gets(line) | | This call seems to trigger the console echo. I am not sure why though. | |
| +- 1231 RE: console echo problem zesty [2004/07/06-15:42] |
| | Response : Email : Edit : Delete |
|
| I
also have also noted this problem. Does anyone have a solution? I am
trying to use | netcat to send and receive text to the AIBO and I always receive the input text repeated | back, then the response. | | >While around with the DSNLookUp demo of the samples, I noticed that every time | >one types an address in the console AIBO responds with an echo. For example if | >you type 123.123.123.123 at the console prompt this is what you would see: | > | >DNSLookUp> 112233..112233..112233..112233 | > | >Now I have look at the code heads and tails and I cant see anywhere an echo | >could be generating unless this is somehow a much lower level bug. Anyone have | >seen this before? | |
| +- 1235 RE: console echo problem TodboT [2004/07/08-01:57] |
| | Response : Edit : Delete |
|
| It
looks like you need to turn off local echo in your terminal program. | |
| +- 1242 RE: console echo problem zesty [2004/07/09-14:10] |
| | Response : Edit : Delete |
|
| >It
looks like you need to turn off local echo in your terminal program. | | No. It doesn't echo everything. Just whenever you do a gets, getc, or scanf. I need a one- | input-line-in, one-output-line-out style interface, and this is not possible as the first line | out is always what I just entered in. It seems to occur in all OPEN-R samples including | the SensorObserver7 and DNSLookup samples, but this is not a usual feature of C++. | | Has anyone else noted this and fixed it somehow? | |
| |- 1243 RE: console echo problem Kuukai [2004/07/09-20:04] |
| | Response : Edit : Delete |
|
| I
noticed WinXp telnet seems local echo by each character. | | I'm using Tera Term. | http://hp.vector.co.jp/authors/VA002416/teraterm.html | It works fine. | | And Win2000 telnet works fine, too. (no local echo) | | Linux telnet works fine, too. | | >>It looks like you need to turn off local echo in your terminal program. | > | >No. It doesn't echo everything. Just whenever you do a gets, getc, or scanf. | I need a one- | >input-line-in, one-output-line-out style interface, and this is not possible | as the first line | >out is always what I just entered in. It seems to occur in all OPEN-R samples | including | >the SensorObserver7 and DNSLookup samples, but this is not a usual feature of | C++. | > | >Has anyone else noted this and fixed it somehow? | |
| +- 1244 RE: console echo problem TodboT [2004/07/10-01:43] |
| Response : Edit : Delete |
|
zesty, It sounds like you want to have a program on your PC communicating with AIBO via port 5900, which is mainly used for debugging. Instead, you might want to do this communication over a different port. Take a look at the EchoServer sample. You can prevent the server echo in this sample. |
| V 1236 Detail information about YVU color space in Aibo needed! wangy22 [2004/07/08-13:07] |
| | Response : Edit : Delete |
|
| I
want to use SDL (Simple Directmedia Layer) to write a program that
display the | video stream from AIBO. It supports YVU color space. | | I've tried transfer the YVU data from AIBO directly to the SDL function, but all | I get is a mess. | | Can anyone give me more information about the color space in AIBO? | | Related functions in SDL are: | | http://sdldoc.csn.ul.ie/sdlcreateyuvoverlay.php | SDL_CreateYUVOverlay $B!#%'(B Create a YUV video overlay | | http://sdldoc.csn.ul.ie/sdllockyuvoverlay.php | SDL_LockYUVOverlay $B!#%'(B Lock an overlay | | http://sdldoc.csn.ul.ie/sdlunlockyuvoverlay.php | SDL_UnlockYUVOverlay $B!#%'(B Unlock an overlay | | http://sdldoc.csn.ul.ie/sdldisplayyuvoverlay.php | SDL_DisplayYUVOverlay $B!#%'(B Blit the overlay to the display | | http://sdldoc.csn.ul.ie/sdlfreeyuvoverlay.php | SDL_FreeYUVOverlay $B!#%'(B Free a YUV video overlay | | Any help would be deeply appreciated. | | Yu | |
| +- 1237 RE:Detail information about YVU color space in Aibo needed! wangy22 [2004/07/08-13:10] |
| | Response : Edit : Delete |
|
| In
SDL, it has 5 YVU formats supported, which one is for AIBO? | | #define SDL_YV12_OVERLAY 0x32315659 /* Planar mode: Y + V + U */ | #define SDL_IYUV_OVERLAY 0x56555949 /* Planar mode: Y + U + V */ | #define SDL_YUY2_OVERLAY 0x32595559 /* Packed mode: Y0+U0+Y1+V0 */ | #define SDL_UYVY_OVERLAY 0x59565955 /* Packed mode: U0+Y0+V0+Y1 */ | #define SDL_YVYU_OVERLAY 0x55595659 /* Packed mode: Y0+V0+Y1+U0 */ | |
| +- 1241 RE:Detail information about YVU color space in Aibo needed! AiboPet [2004/07/09-02:46] |
| Response : HomePage : Edit : Delete |
|
>In
SDL, it has 5 YVU formats supported, which one is for AIBO? The AIBO encoding is unique. In lores mode there are 3 bytes for each pixel (Y+V+U, sometimes called Y Cr Cb). In hires mode there are 6 bytes for each 4 pixels (4 Y pixels, V U). >#define SDL_YV12_OVERLAY 0x32315659 /* Planar mode: Y + V + U */ YV12 comes the closest, however you need to use the hires image data, and mix in the 4 hires Y planes into one. IMHO: it is easier to do the AIBO conversion yourself (sample code provided for that and I have a version for hires color). Convert from YUV/YVU to RGB and then draw the RGB image. |
| V 1238 Your Programs kingdonk [2004/07/08-18:14] |
| | Response : HomePage : Email : Edit : Delete |
|
| Can
every one give details about their programs that they are creating, and | specify if its for the ERS-7 or erlier models. | It may help to stop more than one person creating the same type of programs, | and allows people to see what is being developed and can get help from other | developers on simular projects. | |
| +- 1240 RE:Your Programs ejt [2004/07/09-01:34] |
| Response : HomePage : Email : Edit : Delete |
|
It's
been a few months since our last release, so I'll mention our Tekkotsu project, since your comment is one of the main reasons why it got started: http://www.tekkotsu.org/ End of the academic year delayed the 2.2 release, but I'm hoping to have it out mid-month now that I'm working as a full-time staff researcher. I'm helping lead a team project for the PA Governor's School for the Sciences, it looks like this year's group is going to do some kind of maze solver, hopefully with mapping. Other current projects include precision kinematics using the ROBOOP library, some advanced vision processing filters, integrated support for multi-robot communication, desktop simulation, and including user-submitted patches for sound input processing and a first-person-shooter game style interface for telepresence. Not all of that is going to be finished for the next release though ;) (The article that 2004/07/09-01:32 posted, was modified by Author) |
| V 1239 I have some trouble in using TCPGateway robottj [2004/07/08-22:36] |
| Response : Email : Edit : Delete |
|
I
have some trouble in using TCPGateway.The following is my .cfg
folders,which is using two robots' communication. hostgw.cfg: # Proxy Service port host !ROBOT_PROXY 59001 192.168.0.100 !ROBOT_PROXY 59001 192.168.0.101 # OPEN-R Service Name port host TCPGateway.SendString1.char.S 59002 192.168.0.100 TCPGateway.Send1.char.S 59004 192.168.0.101 TCPGateway.ReceiveString2.char.O 59003 192.168.0.100 TCPGateway.Receive2.char.O 59005 192.168.0.101 robotgw.cfg(1): TCPGateway.Proxy.AperiosMessage.P 59001 TCPGateway.Receive1.char.S 59004 TCPGateway.Send2.char.O 59005 robotgw.cfg(2): TCPGateway.Proxy.AperiosMessage.P 59001 TCPGateway.ReceiveString1.char.O 59002 TCPGateway.SendString2.char.S 59003 Can you tell me what'wrong? Or tell me how to use TCPGateway? Thanks . |
| V 1223 How to define color detect table? aiboKang [2004/07/02-03:59] |
| | Response : Edit : Delete |
|
| Hi, | To make ers-7 identify color based objects, it's necessary to define color | detect table. Color detect table for pink is provided in open-r sample, but | how is the color detect table defined? | If to get color detect table for other colors such as red, blue, sky-blue, | orange and so on, what and how should I do ? | Another question, How are ers-7 camera data defined? what kind of camera data | format is used ? How can ers-7 identify objects with different colors? how | many colors can ers-7 identify? | | Thanks very much , | kandar | | |
| |- 1225 RE:How to define color detect table? AiboPet [2004/07/03-00:42] |
| | Response : HomePage : Edit : Delete |
|
| >Hi, | > Color detect table for pink is provided in open-r sample, but | >how is the color detect table defined? | From the "Level2ReferenceGuide_E.pdf" | | "OCdtInfo Description | In the color detection table, Y (a luminance signal) is divided into 32 | segments, and | Crmax, Crmin, Cbmax and Cbmin are specified for each segment of Y. | The values of Cr and Cb are offset binary ranging from 0x0 to 0xff. | " | For setting other color ranges, see the samples (which have dumb ranges IMHO) | | >If to get color detect table for other colors such as red, blue, sky-blue, | >orange and so on, what and how should I do ? | | There are several ways of doing this. | One way is to take the TR.CT file (which the regular AiboMind uses) and | extract the values from it (I posted sample code a while ago, but it may be | hard to find). This works great for any of the canned colors: Pink, Red, | Flesh, Blue, Green | | Another was is to mathematically calculate ranges based on a central RGB | color. This approach does not work as well as expected because the camera | color response. | | The best way is to capture images with the actual ERS-7 camera, in different | lighting conditions. This has been done in the past for RoboCup dogs. | | >Another question, How are ers-7 camera data defined? what kind of camera data | >format is used ? | See the manual and any data structures starting with "OFbk" (the camera data | is processed through a multi-resolution "filter bank") | | > How can ers-7 identify objects with different colors? how many colors can | ers-7 identify? | See above. | This is done at a hardware level (and is very fast). There are 8 distinct | color ranges. | There is a general color frequency value (1 byte for each of the 8 colors). | The more detailed CDT image has one byte for each "pixel". For the ERS-7 this | is 104x80 bytes. Each byte has 1 bit for each color. | | If you want any more detail than that, or another color detection technique, | you will have to look at the raw image data in software (much slower than | using the hardware) | |
| +- 1234 RE:How to define color detect table? openrsupport@(Administrator) [2004/07/08-01:48] |
| Response : Edit : Delete |
|
Please
also read Training Course "3. Receive data from a system object" from the OPEN-R SDK University web page. https://openr.aibo.com/openr/eng/perm/openrsdk/university2.php4 It explains how to use AIBO's camera data and color detection tables. However, it doesn't explain how to make the color detection table data. AIBO SDE Support |
| V 1228 Compiled suceessfullly but cannot run robottj [2004/07/05-12:28] |
| | Response : Email : Edit : Delete |
|
| I've met a problem when I programmed with OPEN_R_SDK. | | My program has been compiled sucessfully with no warning and error. | **************************************************************** | CXXFLAGS= \ | -O2 \ | -g \ | -I. \ | -I$(OPENRSDK_ROOT)/OPEN_R/include/R4000 \ | -I$(OPENRSDK_ROOT)/OPEN_R/include \ | -Wall \ | -ansi | **************************************************************************** | But when we run the program ,the system told us there is syntax error. | *********************************************************** | $ RP_OPEN_R/bin/start-rp-openr | [pid:3108,msqid:2,oid:0x00000002] oserviceManager | [pid:2440,msqid:3,oid:0x00000003] tcpGateway | MS/OPEN-R/MW/OBJS/Mediator.BIN: 1: Syntax error: "(" unexpected | OID of MS/OPEN-R/MW/OBJS/Mediator.BIN is invalid. | [oid:00000002,prio:1] OServiceManager::Connect() : Mediator.ReceiveString.char.O | NOT FOUND | [oid:00000002,prio:1] OServiceManager::Connect() : Mediator.SendString.char.S NO | T FOUND | ***************************************************************************** | Long for your help. | robottj | |
| +- 1233 RE:Compiled suceessfullly but cannot run openrsupport@(Administrator) [2004/07/08-01:36] |
| Response : Edit : Delete |
|
Please
read Training Course "7.Debugging by gdb using Remote Processing OPEN-R" in the OPEN-R SDK University web page. https://openr.aibo.com/openr/eng/perm/openrsdk/university2.php4 It explains how to use Remote Processing OPEN-R. The binary that executes on the PC side should be an Intel binary. So it should not include mips R4000 headers, > -I$(OPENRSDK_ROOT)/OPEN_R/include/R4000 \ > -I$(OPENRSDK_ROOT)/OPEN_R/include \ and the compiler should be native (for Intel g++) CXX=g++ The binary that executes on AIBO should be a mips binary. So it should include the mips R4000 headers, and the compiler should be the mips cross compiler CXX=$(OPENRSDK_ROOT)/bin/mipsel-linux-g++ Those are the main difference. CP= INSTALLDIR= STRIP= MKBIN= STUBGEN= MKBINFLAGS= LIBS= CXXFLAGS= OCF= are also different from the PC's binary and AIBO's binary. In STEP2 of Training Course 7, it is explained how to modify your makefile to make each binary. Thank you, AIBO SDE Support |
| V 1232 MS Image patch for AIBO Remote Framework has been released. openrsupport@(Administrator) [2004/07/08-01:26] |
| Response : Edit : Delete |
|
A
Memory Stick Image patch for AIBO Remote Framework has been released.
The problem that sample contents created by AIBO Motion Editor cannot be played with AIBO will be fixed. AIBO SDE Support |
| V 1230 FYI: RCODE and YART, BallTracking AiboPet [2004/07/06-03:20] |
| Response : HomePage : Edit : Delete |
|
A
general reply that may be useful to others looking into RCODE YART is a simple tool for editing parts of the R-CODE.R script. It comes with a simple default personality (the initial R-CODE.R on the YART7 installation). Everything you do with YART can be done with a text editor. There are many combinations of using YART and direct RCODE together. First, use YART7 as-is. It will edit the R-CODE.R file for you, and eliminate the chances of "typos" Next, look at the R-CODE.R file that was edited by YART. You can learn from this, and add more structure if you like. For example, YART will let you add simple things, but if you want more random conditionals, you need to edit the file. Finally, you can add additional YART features by adding the proper "//{" ... "//}" comments to the file. This is advanced, but also very simple, especially if you want your personality to be customized by others. ------ > Can I make AIBO swing his head to look for the ball like BallTracking7 in OPEN-R sample? To reproduce BallTracking7, you just need one line: PLAY ACTION TRACK_HEAD PINK_BALL There are various "SEARCH" actions for finding the ball in the first place. The only target supported is "PINK_BALL" (notice it is all upper case) The "Pink_Ball" variable is something else (notice it is mixed case) Typically your main program will wait in a loop for "Pink_Ball" to be non-zero, then you can call "PLAY ACTION TRACK_HEAD PINK_BALL" to have AIBO track it. |
| V 1227 Where can I find more sample codes for R-CODE ? jastw [2004/07/04-04:11] |
| | Response : Email : Edit : Delete |
|
| Where can I find more sample codes for R-CODE? | | Thank you! | |
| +- 1229 RE:Where can I find more sample codes for R-CODE ? AiboPet [2004/07/06-03:16] |
| Response : HomePage : Edit : Delete |
|
>Where
can I find more sample codes for R-CODE? Several suggestions: There are a very few simple ones in the "Redist7\Samples" folder with the ERS- 7 version of RCODE. A simple skeleton program with a main loop and a number of handlers is provided with YART. http://www.aibohack.com/rcode/yart.htm (ERS-2x0/31x, RCode and RCodePlus) http://www.aibohack.com/ers7/yart7.htm (ERS-7, straight RCode) RCodePlus includes a few "RCR_" helper subroutines. Many of the free custom "entertainment" personalities have this same structure (just different reactions, sounds and canned skits) http://www.aibohack.com/index.html#GRID DogsLife is the most complicated RCODE/RCodePlus program written to date. It uses the AibNet pre-processing tools. ERS-2x0/31x only right now. http://www.dogsbodynet.com/dogslife.html http://www.dogsbodynet.com/aibnet.html Finally, for Master Studio owners (2x0/31x), there are some .BE samples provided with it. The R-CODE generated by Master Studio is not human readable, but you it has a few ideas. Like most things, the actual RCODE needed to do a simple task can be 1 or 2 lines of code. The main structure of the application is what makes it harder to have generally usable sample code. If you have a particular area of interest, I or others may have some test code [RCODE is great for trying things out, whether you keep the code or not] |
| V 1226 Author DELETE ***** [2004/07/04-04:07] |
| Response : Edit : Delete |
|
(The
article that 2004/07/03-08:31 posted, was modified by Author) |
| V 1144 (7) Item of Important Attention has been added openrsupport@(Administrator) [2004/06/09-00:44] |
| | Response : Edit : Delete |
|
| (7)
Item of Important Attention has been added. (7) Please turn off the
joint | gain when AIBO is operating and sitting on the Energy Station for an extended | period. | | Thank you, | | AIBO SDE Support | (The article that 2004/06/09-00:44 posted, was modified by Author) | |
| +- 1224 Please use LMS commands instead of GainOff() openrsupport@(Administrator) [2004/07/02-08:22] |
| Response : Edit : Delete |
|
The
7th item of Important Attention has been modified. To set the joint gain off, please use LMS commands instead of GainOff(). Execute LMS commands to set the joint gain values OFF when AIBO is seated on the Energy Station. http://openr.aibo.com/openr/eng/no_perm/notice.php4 FAQ C0003 for AIBO Remote Framekwork has been modified. To set the join gain off, please use two LMS commands instead of CVAIBO:GainOff() CVAIBO:GainOn(), CVAIBO:GainOff() will not be supported in the next release of AIBO Remote Framework. https://openr.aibo.com/openr/eng/perm/rfw/regi_faq3.php4#C0003 Thank you, AIBO SDE Support |
| V 1218 Is there a IDE tool for AIBO programming? jastw [2004/07/01-06:03] |
| | Response : Edit : Delete |
|
| Is there a IDE tool for AIBO programming? | | How can I learn Make and Makefile for compiling the program under OPEN-R? | | Thank you so much | | Sincerely, | | Jason | |
| +- 1222 RE:Is there a IDE tool for AIBO programming? AiboPet [2004/07/02-03:02] |
| Response : HomePage : Edit : Delete |
|
>Is
there a IDE tool for AIBO programming? It depends what you mean by "IDE".... Which platform and which tool is best for you depends on the what you want the robot to do as well as your programming skills. There are several GUI tools. Sony used to sell "Master Studio" ($350 to $450) for the ERS-2x0/31x model AIBOs. It had a flow-chart-like behavior editor and as well as a motion editor. There are free tools: For motion editing, I recommend Skitter http://www.dogsbodynet.com/skitter.html For behavior programming, I recommend that everyone start with YART. It is free and available for all model AIBOS. It relies on RCode/RCodePlus and gives you quick access to most AIBO features. http://www.aibohack.com/rcode/yart.htm [ERS-210/220/31x version] http://www.aibohack.com/ers7/yart7.htm [ERS7 version] After that, you can decide which way to go: more RCODE programming, using AibNet or other tools, or RFW (ERS-7) or the raw power of the OpenR SDK (low level robot control with C++) >How can I learn Make and Makefile for compiling the program under OPEN-R? Here's my advice for getting started: http://www.aibohack.com/openr_sdk/advice.htm FWIW: I strongly recommend everyone start with something easier like RCODE. |
| V 1220 Author DELETE ***** [2004/07/01-21:47] |
| Response : Edit : Delete |
|
(The
article that 2004/07/01-21:38 posted, was modified by Author) |
| V 1085 Timer laurenshobert [2004/05/14-21:55] |
| | Response : Email : Edit : Delete |
|
| I
was wondering if it is possible to create some kind of timer which will
call | a method after a specified period? | |
| +- 1166 RE:Timer SignX [2004/06/14-23:14] |
| | Response : Edit : Delete |
|
| >I
was wondering if it is possible to create some kind of timer which will
call | >a method after a specified period? | | We've created a OPEN-R Timer object. You can connect to this OPEN-R object and | send a TimerRequest. When a certain time has passed, this Timer OPEN-R object | will return a message (event). It works nice! If you want, you can get the | BINARY (and maybe the source too, that I have to ask). | |
| |- 1171 RE:Timer ejt [2004/06/15-03:11] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | >We've
created a OPEN-R Timer object. You can connect to this OPEN-R object and | | >send a TimerRequest. When a certain time has passed, this Timer OPEN-R object | | >will return a message (event). It works nice! If you want, you can get the | | >BINARY (and maybe the source too, that I have to ask). | | | | I'm curious, what's the accuracy of your timers? As in, if you request a timer for X ms, what is the | | variation on the timing of the message that is returned? | | | | I ask because once upon a time OPEN-R messages had a significant amount of lag, but I hear they're | | better now... but I'm curious to have some numbers to back it up... | | | | thanks! | | |
| | +- 1190 RE:Timer SignX [2004/06/23-19:22] |
| | Response : Edit : Delete |
|
| >>We've
created a OPEN-R Timer object. You can connect to this OPEN-R object and | >>send a TimerRequest. When a certain time has passed, this Timer OPEN-R object | >>will return a message (event). It works nice! If you want, you can get the | >>BINARY (and maybe the source too, that I have to ask). | > | >I'm curious, what's the accuracy of your timers? As in, if you request a timer | for X ms, what is the | >variation on the timing of the message that is returned? | > | | We have no data on accuracy or variation, we've used the interal frame number | as time (received from OVirtualRobotComm). One frame is 8 ms. We haven't tested | the object very intensively because the timer didn't had to be very accurate. | | >I ask because once upon a time OPEN-R messages had a significant amount of | lag, but I hear they're | >better now... but I'm curious to have some numbers to back it up... | > | >thanks! | |
| +- 1219 RE:Timer laurenshobert [2004/07/01-20:13] |
| Response : Email : Edit : Delete |
|
>We've
created a OPEN-R Timer object. You can connect to this OPEN-R object and >send a TimerRequest. When a certain time has passed, this Timer OPEN-R object >will return a message (event). It works nice! If you want, you can get the >BINARY (and maybe the source too, that I have to ask). I'm very interested in your OPEN-R Timer object. You can contact me at: l(dot)h (dot)x(dot)hobert(at)student(dot)utwente(dot)nl |
| V 1217 Where to buy ERA-201 Wireless LAN Card now? wangy22 [2004/07/01-03:45] |
| Response : Edit : Delete |
|
Hi, I want to buy a new wireless card for AIBO ERS-210. I checked the online store of AIBO US and found that ERA-201 Wireless LAN Card is not being sold any longer. Does anyone know any other source I can purchase a new wireless card for it? Thanks a lot! Yu |
| V 1198 Commands for the ERS 210 djtommyc [2004/06/25-06:19] |
| | Response : Email : Edit : Delete |
|
| IS
there anywhere I can find a list of all Rcode commands for the ERS 210?
What | I mean by commands are things like "WALK" and "MOVE_HEAD". | | thanks, | tom | |
| +- 1200 RE:Commands for the ERS 210 TodboT [2004/06/25-08:24] |
| | Response : Edit : Delete |
|
| Try this: http://www.us.aibo.com/rcode/rcode12ref.txt | |
| +- 1203 RE:Commands for the ERS 210 AiboPet [2004/06/27-00:14] |
| | Response : HomePage : Edit : Delete |
|
| >Try
this: http://www.us.aibo.com/rcode/rcode12ref.txt | That describes the basic syntax. | | > ...are things like "WALK" and "MOVE_HEAD". | | Those are special cases that start with "PLAY ACTION". | Check out the ACTION.AMS file on the memory stick for the built-in actions | (\OPEN-R\APP\PC\AMS folder) | BTW: there was an older text document describing these, but you should be able | to figure most of them out from the text file | | NOTE: In addition if there are ACTION.EXT and ACTION.CFG in that same | directory, they can be used as well with "PLAY ACTION ????" | [by convention: ACTION.AMS = built in system things, ACTION.EXT = add ons like | RCodePlus library of hundreds of canned actions, ACTION.CFG = personality | specific, generated by Master Studio or YART or other tools] | |
| +- 1206 RE:Commands for the ERS 210 djtommyc [2004/06/29-04:40] |
| | Response : Email : Edit : Delete |
|
| >>Try
this: http://www.us.aibo.com/rcode/rcode12ref.txt | >That describes the basic syntax. | > | >> ...are things like "WALK" and "MOVE_HEAD". | > | >Those are special cases that start with "PLAY ACTION". | >Check out the ACTION.AMS file on the memory stick for the built-in actions | >(\OPEN-R\APP\PC\AMS folder) | >BTW: there was an older text document describing these, but you should be able | >to figure most of them out from the text file | > | >NOTE: In addition if there are ACTION.EXT and ACTION.CFG in that same | >directory, they can be used as well with "PLAY ACTION ????" | >[by convention: ACTION.AMS = built in system things, ACTION.EXT = add ons like | >RCodePlus library of hundreds of canned actions, ACTION.CFG = personality | >specific, generated by Master Studio or YART or other tools] | | That helps. Is there a place where I can find the actions for things like "move | front left leg clockwise 30 degrees?" Also, is there somewhere it explains the | paramiters for the commands? | |
| +- 1216 RE:Commands for the ERS 210 AiboPet [2004/07/01-03:28] |
| Response : HomePage : Edit : Delete |
|
>That
helps. Is there a place where I can find the actions for things like "move >front left leg clockwise 30 degrees?" The built-in actions do not have that level of control. The only thing that comes close is the MOVE_HEAD (where you can specify 2 angles) The RCodePlus AP_MOVE_JOINTS feature gives you detailed control over each joint. http://www.aibohack.com/rcode/rcp_motion.htm > Also, is there somewhere it explains the >paramiters for the commands? There was an old .TXT file on the website which I can't find right now. The RCODE reference gives the pertinent details: ----- PLAY:ACTION:<action>[:<arg1>:<arg2>:<arg3>] ... Argument Pattern <horizontal angle>:<distance> // WALK, KICK, TOUCH, etc. <horizontal angle> // TURN, etc. <horizontal angle>:<vertical angle> // MOVE_HEAD, etc. <TARGET> // SEARCH, etc. Angle unit is degrees Distance unit is millimeters Currently, only PINK_BALL can be specified as <TARGET>. |
| V 1193 The license of the file structure of the motion files. wangy22 [2004/06/24-05:20] |
| | Response : Edit : Delete |
|
| I am
wondering if it is possible to use the file structure of the motion
files | created by Master Studio freely in open source programming(say GPL)? | | How about the code of the motion manager in the OPEN_R Samples and the existing | motions? | |
| |- 1197 RE:The license of the file structure of the motion files. AiboPet [2004/06/25-04:38] |
| | Response : Edit : Delete |
|
| I'd
like to hear an official reply. I-Am-Not-A-Lawyer, but here is my
advice: | >I am wondering if it is possible to use the file structure of the motion files | >created by Master Studio freely in open source programming(say GPL)? | | The factual information inside the SDK is not a trade-secret, so you can use | the format information to do anything you want (ie. you agreed to an "EULA" | not an "NDA") | However: The EULA limits the distribution of Sony copyrighted material (ie. | sample implementations, documentation files, headers etc) | | >How about the code of the motion manager in the OPEN_R Samples and the | existing motions? | From EULA: "3.3 Assignment. You may not assign or transfer any of the | Development Kit to a third party or allow a third party to use the same" | | Placing parts of the Open-R SDK into a GPL project falls under that case. | | (IANAL) | |
| +- 1207 RE:The license of the file structure of the motion files. openrsupport@(Administrator) [2004/06/29-08:08] |
| | Response : Edit : Delete |
|
| Hello, | | >I am wondering if it is possible to use the file structure of the motion files | >created by Master Studio freely in open source programming(say GPL)? | | Contents created with AIBO Master Studio are not 'open source', but | you can distribute them. See 4.RIGHT in the END-USER LICENSE AGREEMENT | for AIBO Master Studio. | | >How about the code of the motion manager in the OPEN_R Samples and the existing | >motions? | | See FAQ #A0002 and #A0012. You can distribute software and motion files | created with the OPEN-R SDK, but commercial use is forbidden. | | Did this answer your questions? If not, please reply with a more | detailed question. Telling us what your intentions are would | be helpful for us to answer your question more appropriately. | | Thanks, | | AIBO SDE Support | |
| +- 1208 RE:The license of the file structure of the motion files. wangy22 [2004/06/29-09:39] |
| | Response : Edit : Delete |
|
| >Hello, | > | >>I am wondering if it is possible to use the file structure of the motion files | >>created by Master Studio freely in open source programming(say GPL)? | > | >Contents created with AIBO Master Studio are not 'open source', but | >you can distribute them. See 4.RIGHT in the END-USER LICENSE AGREEMENT | >for AIBO Master Studio. | > | >>How about the code of the motion manager in the OPEN_R Samples and the existing | >>motions? | > | >See FAQ #A0002 and #A0012. You can distribute software and motion files | >created with the OPEN-R SDK, but commercial use is forbidden. | > | >Did this answer your questions? If not, please reply with a more | >detailed question. Telling us what your intentions are would | >be helpful for us to answer your question more appropriately. | > | >Thanks, | > | >AIBO SDE Support | | Hi, | | Currently I am developing an AIBO Remote Control, for Linux, as an alternative | to the existing controlling software for Windows. The program is wanted to be | released under GPL license. Since my programming skill is not advanced, so I am | wondering if I can use some codes from OPENR-SDK, OPENR-SDK Samples, as well as | the pre-defined motions in MoNet in my program? | | Now I am designing my own file structure for storing the motions, as what | mentioned above that the existing licenses in OPENR SDK conflicts with GPL. | |
| +- 1215 RE:The license of the file structure of the motion files. AiboPet [2004/07/01-03:17] |
| Response : HomePage : Edit : Delete |
|
Commentary
(IANAL) >... the existing licenses in OPENR SDK conflicts with GPL. Of course the standard GPL does not restrict commercial use. GPL's version of free is like "free speech" Open-R SDK EULA version of free is like "free beer" My suggestion would be to release it under both! (a combined licence) Both the GPL *and* the Open-R SDK EULA. All derived work would remain freely available (in the GPL "free speech" sense) for non-commercial use ("free beer") assuming no other conflicts with Sony's terms. |
| V 1210 obstacle detection ERS-7 Gilles [2004/06/29-18:18] |
| | Response : Edit : Delete |
|
| With
R-CODE is it possible for ERS-7 to "see" walls, etc. with its head
sensor ? | In R-CODE doc, we found "Distance" and "Distance_cliff" instructions but | without details about. | | I tried the following code (for a "go forward" voice command) and Aibo walks | well 0.5 meter but doesn't detect wall or obstacle : | | FOR:i:1:10 | PLAY ACTION WALK 0 50 | WAIT | IF:Distance:<:50:THEN | PLAY ACTION GO RIGHT 90 | WAIT | END IF | NEXT | | Someone has an idea about this problem ? | (The article that 2004/06/29-18:17 posted, was modified by Author) | |
| +- 1211 RE:obstacle detection ERS-7 Kuukai [2004/06/29-20:28] |
| | Response : Edit : Delete |
|
| Try this program with wireless LAN | | FOR:i:1:10 | PRINT "A" | PLAY:ACTION:WALK:0:50 | WAIT | VDUMP Distance | IF:Distance:<:150:THEN | PRINT "B" | PLAY:ACTION:TURN:90 | WAIT | ENDIF | NEXT | | >With R-CODE is it possible for ERS-7 to "see" walls, etc. with its head | sensor ? | >In R-CODE doc, we found "Distance" and "Distance_cliff" instructions but | >without details about. | > | >I tried the following code (for a "go forward" voice command) and Aibo walks | >well 0.5 meter but doesn't detect wall or obstacle : | > | >FOR:i:1:10 | >PLAY ACTION WALK 0 50 | >WAIT | > IF:Distance:<:50:THEN | > PLAY ACTION GO RIGHT 90 | > WAIT | > END IF | >NEXT | > | >Someone has an idea about this problem ? | >(The article that 2004/06/29-18:17 posted, was modified by Author) | |
| +- 1212 RE:obstacle detection ERS-7 Gilou94 [2004/06/30-04:17] |
| | Response : Edit : Delete |
|
| OK
it now works fine with Distance set 150 or 100 (the right code for turn
is yours, I made a mistake | while writing the message). | | Thanks | | >Try this program with wireless LAN | > | >FOR:i:1:10 | >PRINT "A" | >PLAY:ACTION:WALK:0:50 | >WAIT | >VDUMP Distance | >IF:Distance:<:150:THEN | >PRINT "B" | >PLAY:ACTION:TURN:90 | >WAIT | >ENDIF | >NEXT | > | >>With R-CODE is it possible for ERS-7 to "see" walls, etc. with its head | >sensor ? | >>In R-CODE doc, we found "Distance" and "Distance_cliff" instructions but | >>without details about. | >> | >>I tried the following code (for a "go forward" voice command) and Aibo walks | >>well 0.5 meter but doesn't detect wall or obstacle : | >> | >>FOR:i:1:10 | >>PLAY ACTION WALK 0 50 | >>WAIT | >> IF:Distance:<:50:THEN | >> PLAY ACTION GO RIGHT 90 | >> WAIT | >> END IF | >>NEXT | >> | >>Someone has an idea about this problem ? | >>(The article that 2004/06/29-18:17 posted, was modified by Author) | |
| +- 1214 RE:obstacle detection ERS-7 AiboPet [2004/07/01-03:15] |
| Response : HomePage : Edit : Delete |
|
>FOR:i:1:10 >PLAY:ACTION:WALK:0:50 >WAIT >IF:Distance:<:150:THEN .... Another approach is not to use "WAIT" and have him walk a long distance ("PLAY ACTION WALK 0 10000", 10 meters) Then while he is walking, do the "Distance" check and you can stop him walking ("PLAY ACTION STOP_WALK") when he gets too close to something. This should make his walking more smooth and reactions more immediate |
| V 1213 Important notice about powering off AIBO using OPEN-R::Fatal openrsupport@(Administrator) [2004/06/30-08:30] |
| Response : Edit : Delete |
|
In
OPEN-R SDK 1.1.5-r2, the battery plunger is locked during startup, shutdown, and shutdown by system errors (e.g. detecting excessive current, etc.) to protect the EEPROM. Calling OPENR:Fatal() causes the same situation as when you forcibly remove the battery while AIBO is operating. If you can avoid doing so, please don't use OPENR:Fatal() to power off AIBO, especially in cases where the battery plunger is locked (you cannot remove the battery manually). We are going to release OPEN-R SDK 1.1.5-r3 which prevents AIBO from powering off while writing data to the EEPROM, even if OPENR:Fatal() is called. Thanks, AIBO SDE Support |
| V 1209 Can anyone provide detail about OPENR::SetJointGain()? wangy22 [2004/06/29-09:59] |
| Response : Edit : Delete |
|
How does those PID coefficient work? What are the ranges for gain and shift? Any examples are appreiciated! Thanks! |
| V 1194 ORB::AGR_MS_GetDataBlockFromEEPROM(11) FAILED 2 RETRY... wangy22 [2004/06/24-08:16] |
| | Response : Edit : Delete |
|
| I
recently got another AIBO Dog, which is an older model than ERS-210A. I
tested | my program on it and everything works, except that I found the following error | messages from AIBO Console: | | $ telnet 192.168.0.213 59000 | Trying 192.168.0.213... | Connected to 192.168.0.213. | Escape character is '^]'. | [oid:0x80000040] /MS/OPEN-R/SYSTEM/OBJS/EMGCYMON.BIN | [oid:0x80000041] /MS/OPEN-R/SYSTEM/OBJS/NETCONFS.BIN | [oid:0x80000042] /MS/OPEN-R/SYSTEM/OBJS/ANTTCPIO.BIN | [oid:0x80000043] /MS/OPEN-R/SYSTEM/OBJS/HOOKACT.BIN | [oid:80000037,prio:1] OrbInterfaceProxy::AGR_MS_GetDataBlockToEEPROM() : Invalid | data block number[226] | [oid:80000037,prio:1] ORB::AGR_MS_GetDataBlockFromEEPROM(11) FAILED 2 RETRY... | [oid:80000037,prio:1] OrbInterfaceProxy::AGR_MS_GetDataBlockToEEPROM() : Invalid | data block number[226] | [oid:80000037,prio:1] OrbInterfaceProxy::AGR_MS_GetDataBlockToEEPROM() : Invalid | data block number[226] | [oid:80000037,prio:1] OrbInterfaceProxy::AGR_MS_GetDataBlockToEEPROM() : Invalid | data block number[226] | [oid:80000037,prio:1] OrbInterfaceProxy::AGR_MS_GetDataBlockToEEPROM() : Invalid | data block number[226] | [oid:80000037,prio:1] OrbInterfaceProxy::AGR_MS_GetDataBlockToEEPROM() : Invalid | data block number[226] | [oid:80000037,prio:1] ORB::AGR_MS_GetDataBlockFromEEPROM(11) FAILED AGAIN 2 | [oid:80000037,prio:1] AGRMSDriver::GetDataBlock() : | ORB::AGR_MS_GetDataBlockFromEEPROM() FAILED 2[oid:80000037,prio:1] ... | ... | [oid:80000037,prio:1] ORB::AGR_MS_GetDataBlockFromEEPROM(3) FAILED AGAIN 2 | [oid:80000037,prio:1] AGRMSDriver::GetDataBlock() : | ORB::AGR_MS_GetDataBlockFromEEPROM() FAILED 2[RobotDesign:ERS-210,odrexecMODE3] | | Why is that? | |
| +- 1205 RE:ORB::AGR_MS_GetDataBlockFromEEPROM(11) FAILED 2 RETRY... openrsupport@(Administrator) [2004/06/29-02:03] |
| Response : Edit : Delete |
|
These messages are normal. If you run the OPEN-R SDK with an ERS-210 (older model than the ERS-210A), these messages are shown from the WLAN console. The EEPROM data block number for the OPEN-R SDK is old, so the OPEN-R SDK will automatically fix these invalid data block numbers on the first boot. You will not see these messages after the 2nd boot. AIBO SDE Support |
| V 1202 Walking code... ssacjin [2004/06/26-22:17] |
| | Response : Email : Edit : Delete |
|
| I'm
studying OPEN-R code with ERS-210. I really want to make it walk. | I think that each step I should control the motors, so I need the angle of | joints. Where can i get the angles..?? | |
| +- 1204 RE:Walking code... AiboPet [2004/06/27-00:23] |
| Response : HomePage : Edit : Delete |
|
>I'm
studying OPEN-R code with ERS-210. I really want to make it walk. If you want a direct solution, then I suggest looking at the SoccerLion sample and the "OMWares" library. This contains a middle-ware object that does the walking for you in several different styles. For the knee walk, I recommend looking at the CMU CMPack walking code [from Robocup] A remote control program is available for exploring the different walk styles http://www.aibohack.com/openr_sdk/walkbench.htm [see demo movie] There are other alternatives that have canned walking code (including RCODE) If instead you want to do it all yourself, and implement a walking algorithm, my first advice is -- don't -- it is tricky, has been done before and is not a good first project. Once again look at the RoboCup documentation. The old UNSW team had a very good description of how to implement a walking algorithm, but it glosses over many of the details and endless hours of tweeking to get a good walking algorithm. >I think that each step I should control the motors, so I need the angle of >joints. Where can i get the angles..?? See the basic samples for how to move the joints/motors. FWIW: Getting the current angles of the joints is generally not interesting (except for strange cases). Figuring out what the future angles are supposed to be and how fast they should move is the hard part... |
| V 1201 Technical Specs ejt [2004/06/26-02:03] |
| Response : HomePage : Email : Edit : Delete |
|
I'm
in the process of setting up a kinematics library (I settled on using
the ROBOOP package if anyone is interested) This library also has facilities for computing dynamics, which could let us develop some very cool programs, but I need some additional information regarding each of the motors. I wonder if someone at Sony would be willing to publish the following information for each of the links/ motors: m mass of the link cx center of mass along axis x cy center of mass along axis y cz center of mass along axis z Ixx element xx of the inertia tensor matrix Ixy element xy of the inertia tensor matrix Ixz element xz of the inertia tensor matrix Iyy element yy of the inertia tensor matrix Iyz element yz of the inertia tensor matrix Izz element zz of the inertia tensor matrix Im motor rotor inertia Gr motor gear ratio B motor viscous friction coefficient Cf motor Coulomb friction coefficient PS I found the diagrams in the documentation very handy for making precise on-screen measurements of points of interest, but it would have been more handy if the file had not been password protected from being opened in graphics programs (i.e. Illustrator) |
| V 1192 Volume (or sensitivity) of microphone is too low. wangy22 [2004/06/24-02:44] |
| | Response : Edit : Delete |
|
| I've
just implemented audio streaming from AIBO to PC via TCP/IP connection. | | Now the problem is that in order to clearly hear the sound retrieved from AIBO, | I have to speak to the microphone in a very short distance. If I speak to the | microphone with a distance longer than about 40cm, I hear no sound feedback. | | I searched the OPEN_R programming references and found no API adjusting the | volume of the microphone, so I need some help. | | Or do I have to program the sound to be louder by myself? How? I use ESound | Daemon under Linux. | | Thanks a lot! | |
| +- 1196 RE:Volume (or sensitivity) of microphone is too low. AiboPet [2004/06/25-04:29] |
| | Response : Edit : Delete |
|
| > ... | >I have to speak to the microphone in a very short distance. If I speak to the | >microphone with a distance longer than about 40cm, I hear no sound feedback. | | What model AIBO ? | Are you setting the microphone mode? (Omni directional or Uni-Directional?) | Omni-directional will increase the angular range. | Those are the only settings that do anything AFAIK (I believe ALC is always on) | | Also look at the streaming audio samples, and see if they are coming in with a | reasonable volume (ie. it could be the playback) | | ---- | If you have a 210/220 and access to a Windows PC, I suggest you try AiboRemote | http://www.aibohack.com/210/remote.htm | Check the "Audio Enabled" option and "Complex" User interface | Anything heard by AIBO will be played on the PC speakers (may cause feedback) | There are also "uni" and "omni" buttons to switch the microphone modes | | This doesn't use the OpenR SDK, but will give you a good idea of what the | hardware is capable of. | For me this gives me great audio dynamic range and AIBO can hear me 10 feet | away (in a quiet room, this is good enough for PC based voice recognition ala | BrainBo/HomieGate) | |
| +- 1199 RE:Volume (or sensitivity) of microphone is too low. wangy22 [2004/06/25-07:48] |
| Response : Edit : Delete |
|
It's
ERS 210a. I didn't set the audio mode to omni-directional, I will try it later.... what is the default mode? Thanks! >> ... >>I have to speak to the microphone in a very short distance. If I speak to the >>microphone with a distance longer than about 40cm, I hear no sound feedback. > >What model AIBO ? >Are you setting the microphone mode? (Omni directional or Uni-Directional?) >Omni-directional will increase the angular range. >Those are the only settings that do anything AFAIK (I believe ALC is always on) > >Also look at the streaming audio samples, and see if they are coming in with a >reasonable volume (ie. it could be the playback) > >---- >If you have a 210/220 and access to a Windows PC, I suggest you try AiboRemote >http://www.aibohack.com/210/remote.htm >Check the "Audio Enabled" option and "Complex" User interface >Anything heard by AIBO will be played on the PC speakers (may cause feedback) >There are also "uni" and "omni" buttons to switch the microphone modes > >This doesn't use the OpenR SDK, but will give you a good idea of what the >hardware is capable of. >For me this gives me great audio dynamic range and AIBO can hear me 10 feet >away (in a quiet room, this is good enough for PC based voice recognition ala >BrainBo/HomieGate) |
| V 1195 Aibo Remote Framework on .NET susanto [2004/06/24-14:03] |
| Response : Edit : Delete |
|
Hi, Will Aibo Sony release AIBO Remote Framework component for .NET? is it possible? I am VB.NET programmer. |
| V 1189 I can't build SimpleNave valxing [2004/06/23-15:28] |
| | Response : Email : Edit : Delete |
|
| I get this error: | | H:\AIBO\AIBORemoteFramework_Beta1\RFW_Sample_src_Beta1 | \SimpleNavi\SimpleNavi.cpp(526) : error C2660: 'D3DXCreateFont' : function | does not take 12 parameters | Error executing cl.exe. | | From the DirectX doc I got this: | | HRESULT D3DXCreateFont( | LPDIRECT3DDEVICE9 pDevice, | HFONT hFont, | LPD3DXFONT *ppFont | ); | | Is SimpleNave using an older D3DX API? | | Thanks, | Valerie. | |
| +- 1191 RE:I can't build SimpleNave TodboT [2004/06/24-01:13] |
| Response : Edit : Delete |
|
I
think you may need to install DirectX 9.0 SDK Update (Summer 2003). |
| V 1179 Remote Framework Server Testing Skrubbzilla [2004/06/17-09:06] |
| | Response : Edit : Delete |
|
| Hi,
I've installed the latest versions of all my drivers and the latest | downloads for the remote framework. I've made a wireless connection to my ERS- | 7 with the appropriate configuration on the memory stick. Now when I run the | BaseClient.exe I can connect and I see the blue lights on the dog's head and I | check the VAIBOServer status and it displays a connection. But after that, no | additional commands seem to do anything. According to the doc's, when I | select "remote" on the BaseClient program, the lights should blink and then I | can play the hello action. After hitting "remote", nothing ever happens (no | sounds, no blinking lights, etc.). Anyone have an idea what I might be doing | wrong? | | I've also tried to develop my own WIN32 program (non-MFC) and I can connect | and disconnect from the dog (lights turn blue on connect and off when I | disconnect), I can ping the dog's IP address, but the WM_VAIBO_CONNECT and | other associated messages never appear for me. I check error codes from the | Connect API calls and no error is reported, but it seems that other API calls | succeed but never actually do anything. Anyone run into this or have some | suggestions? | | Thanks! | |
| |- 1180 RE:Remote Framework Server Testing yorkUSA [2004/06/18-07:03] |
| | Response : Edit : Delete |
|
| I am having this exact same problem. | |
| |- 1181 RE:Remote Framework Server Testing grbolar [2004/06/18-10:40] |
| | Response : Edit : Delete |
|
| Are
you using a firewall software on your PC ? | If so, it may be blocking incoming packets from AIBO. | |
| |- 1182 RE:Remote Framework Server Testing AiboPet [2004/06/19-02:09] |
| | Response : Edit : Delete |
|
| >Hi,
I've installed the latest versions of all my drivers and the latest | >downloads for the remote framework. I've made a wireless connection to my ERS- | >7 with the appropriate configuration on the memory stick. | Some suggestions: | | As mentioned, make sure you don't have a firewall blocking access. | I use ZoneAlarm and the first time VAIBOServer.exe is run it needs to be given | Client and Server access to the WiFi (it uses a lot of different ports) | | > Now when I run the | >BaseClient.exe I can connect and I see the blue lights on the dog's head and I | >check the VAIBOServer status and it displays a connection. | | Good. FWIW: I also recommend RemoteTest, it has many extra features, but | BaseClient is a good place to start. | | > But after that, no | >additional commands seem to do anything. According to the doc's, when I | >select "remote" on the BaseClient program, the lights should blink and then I | >can play the hello action. After hitting "remote", nothing ever happens (no | >sounds, no blinking lights, etc.). Anyone have an idea what I might be doing | >wrong? | | Sounds like you are doing it correctly. | Yes it should respond (with sound and flashing LEDs) when you switch modes. | | When it stops responding, try pinging it just to be sure it is there. | | Are the ears still blue ? blinking ? | (should be solid blue when first connect, blinking blue if in Remote mode, ear | lights off if disconnected) | | >I've also tried to develop my own WIN32 program (non-MFC) and I can connect | >and disconnect from the dog (lights turn blue on connect and off when I | >disconnect), I can ping the dog's IP address, but the WM_VAIBO_CONNECT and | >other associated messages never appear for me. | | You must call CVAIBO::Connect to initiate the process. That part should be | working based on your experience with BaseClient. | | Other advice: | Check that your WiFi connection is reliable (depending on your WiFi client) | An access point can help to facilitate connections better than ad-hoc | Make sure you have a full RFW installation on the stick (ie. it won't fit on | an 8MB stick) | |
| |- 1186 RE:Remote Framework Server Testing openrsupport@(Administrator) [2004/06/23-01:09] |
| | | Response : Edit : Delete |
|
| | Firewall
(or antivirus) software may be blocking network ports in one direction. | | | | PC --> AIBO : Pass | | AIBO --> PC : Blocking | | | | Can you please post your environment? | | | | 1. OS type: | | 2. Firewall software: | | 3. What happens when you execute RemoteTest and push the connect button? | | RemoteTest will show the "AIBO Output" window and will report the data coming | | from AIBO. | | What messages do you get in "AIBO Output"? | | 4. Disable Firewall software and try to get AIBO remote controlled. | | If you succeed, change the security level of your Firewall software, or | | manually open the ports which AIBO uses. | | | | These FAQs may also help you: | | https://openr.aibo.com/openr/eng/perm/rfw/regi_faq3.php4#C0005 | | https://openr.aibo.com/openr/eng/perm/rfw/regi_faq3.php4#D0001 | | (4) If PC applications made by AIBO Remote Framework cannot | | connect to AIBO, but pinging AIBO is successful: | | | | Thank you, | | | | AIBO SDE Support | | (The article that 2004/06/23-01:08 posted, was modified by Author) | | |
| | +- 1188 RE:Remote Framework Server Testing Skrubbzilla [2004/06/23-05:19] |
| | Response : Edit : Delete |
|
| It
seems that the built in Windows XP firewall will block incoming info
from | the AIBO. I turned the XP firewall off and all the demo programs work | correctly. And my Win32 API code works fine to. Thanks everyone! | |
| +- 1187 RE:Remote Framework Server Testing TodboT [2004/06/23-01:33] |
| Response : Edit : Delete |
|
I
think I had the same problem. It wasn't caused by a firewall, but by
some VPN software. After I uninstalled it, everything worked fine. |
| V 1185 Head joints problems ckch [2004/06/22-18:21] |
| Response : Edit : Delete |
|
Hello, When I started up my aibo(ERS-7), the head pan and tilt are fine and stiff, but after playing for 10 seconds or so, they suddenly became loose, and they are free to move, (except for the crane). I tried with my other dog with the same stick, but that doesnt happen. So it doesnt seem to be a program bug. Does anyone know why this could happen? If the hardware is broken, how come the joints were stiff when it started up? Does anyone have a similar problem Thanks very much for reading my message. |
| V 1153 what is LogicalMemoryStickInterface? robottj [2004/06/10-19:24] |
| | Response : Email : Edit : Delete |
|
| hi
,I am doing the vision module of four-legged league and I want to save
the | pictures the robot takes on the MemoryStick.But every time I run the codes ,the | wirelss console occurs the following | error:LogicalMemoryStickInterface::samePage::index(300)original[0]->copy | [ff],and the picture can't be saved to the memory stick.But if I turn the robot | off and then turn it on again,the code works normally.Could anyone tell me what | is wrong with it?Is there anything wrong with my code?Thanks! | |
| +- 1159 RE:what is LogicalMemoryStickInterface? grbolar [2004/06/11-12:05] |
| | Response : Edit : Delete |
|
| If
you have some bad pointer arithmetic causing you to write into the
system | memory area, anything wrong can happen. | Try protected memory. (See #1110) | https://openr.aibo.com/cgi-bin/openr/e_regi/im_trbbs.cgi? | uid=general&df=bbs.dat&prm=TAN&pg=2&no=1110#1110 | |
| +- 1183 RE:what is LogicalMemoryStickInterface? robottj [2004/06/21-23:00] |
| | Response : Edit : Delete |
|
| >If
you have some bad pointer arithmetic causing you to write into the
system | >memory area, anything wrong can happen. | >Try protected memory. (See #1110) | >https://openr.aibo.com/cgi-bin/openr/e_regi/im_trbbs.cgi? | >uid=general&df=bbs.dat&prm=TAN&pg=2&no=1110#1110 | But I run the sample programme--ImageOberver and it occurs the same result.I | don't really understand your meaning.Could you please give me more details? | Thanks! | |
| +- 1184 RE:what is LogicalMemoryStickInterface? grbolar [2004/06/22-16:48] |
| Response : Edit : Delete |
|
>But
I run the sample programme--ImageOberver and it occurs the same result. Then, media defect is possible. Try another memorystick. |
| V 1164 Remote FW : CVAIBO derivation problem + RemoteTest bug chaxel [2004/06/14-20:45] |
| | Response : Email : Edit : Delete |
|
| Hello, | | I would like to use RemoteFW and AIBO ERS_7 for basic C++ learning with my | students. | | For this it will be interrested to derivate CVAIBO. | | class CMyVAIBO: public CVAIBO | { | | ..... | | } | | Because CVAIBO constructor is not the default one, I must provide | | CMyVAIBO::CMyVAIBO(HWND parent) | { | CVAIBO::CVAIBO(parent); | } | | But it couldn't be done because CVAIBO default constructor (the one with no | parameter) is not provided, and in this case VC++ say 'error 2512 no | appropriate defaut constructor available' | | Of course as I have try you can imagine that it will be resolve by modifiying | the CVAIBO.h by adding CVBAIO(); methods declaration ... I can suppose it will | be possible because my code don't use it !!! | Yep OK for the compiler, but not OK for the linker who search for CVAIBO:CVAIBO | even if it is not used in my code ! | One step more you can try to povide CVAIBO::CVAIBO() in you code. | Re-YEP, OK for the linker (with a small warning), but not OK during execution | VAIBOServer is not launch. | | SO if everybody could provide a solution ! Or maybe in the next release of RFW | Sony could provide and export CVAIBO::CVAIBO in the dll in order to permit base | class derivation ... even if it must not be used for proper operation of | VAIBOServer. | | SECOND POINT : Small Bug Report | | in Application RemoteTest (Sample RemoteFW) | | File RemoteTestDlg.Cpp | CRemoteTestDlg::OnVAIBOConnect (...) | | button3DDlg().EnableWindow(); is missing so 3D Button | cannot be accessed | | File 3DDlg.cpp | void C3DDlg::OnBtnLoad() | | the test of Robot design say "Invalid RobotDesign" with | my brand new ERS7 | ... if I put out this test | if (m_robotDesign != DESIGN_ERS7) ... and so on | then all is OK and the 3D ERS7 is animated (when | sensors channel open) as the real one | |
| |- 1167 RE:Remote FW : CVAIBO derivation problem + RemoteTest bug chaxel [2004/06/14-23:28] |
| | Response : Email : Edit : Delete |
|
| Hi, | | Reply to myself concerning constructor | | class CMyVAIBO : public CVAIBO | { | CMyVAIBO (HWND Parent) : CVAIBO (parent) {}; | OthersMethods | | }; | | .... is OK | | Call to base class constructor has to be made inside the derivated class | declaration. | | Bug in RemoteTest remaing active hovewer. | |
| |- 1168 RE:Remote FW : CVAIBO derivation problem + RemoteTest bug AiboPet [2004/06/14-23:30] |
| | Response : Edit : Delete |
|
| >class CMyVAIBO: public CVAIBO | ... | >Because CVAIBO constructor is not the default one, I must provide | > | > CMyVAIBO::CMyVAIBO(HWND parent) | > { | > CVAIBO::CVAIBO(parent); | > } | | Try: | | CMyVAIBO::CMyVAIBO(HWND parent) | : CVAIBO(parent) | { | // put other code here | } | | [that's just C++ syntax, one of the dumber parts of the language IMHO -- | nothing specific to AIBO] | | BTW: I agree it would be nice to have a parameter-less constructor, and the | option of connecting it to an HWND later after construction. Then you wouldn't | be forced to use 'new CVAIBO', and instead could have a regular global | variable. | |
| +- 1178 RE:Remote FW : CVAIBO derivation problem + RemoteTest bug openrsupport@(Administrator) [2004/06/16-02:44] |
| Response : Edit : Delete |
|
Hello, Regarding (1) Parameter-less constructor of CVAIBO, CVAIBO() requires a Windows handle to send Windows Messages from VAIBOServer. We understand that a parameterless constructor is useful in the initialization section. However, CVAIBO() doesn't work without a Windows handle. We don't want to create a new API, such as CVAIBO::SetWindowsHandle(HWND). The reason for this is that Windows Messages come to the client whenever AIBO's status is changed (e.g. battery level changed) even if the client does not send any command to AIBO. We prefer not to make programmers go through the extra work of maintaining a critical section while changing the associated Windows handle. Regarding (2) GetRobotDesign() returns DESIGN_UNKNOWN = 0, This is a bug. The bug will be fixed in the next release. GetRobotDesign() returns the correct AIBO model after calling RequestNetService( SENSOR_ID ); GetRobotDesign() determines the type of your AIBO when it gets the AIBO's CPC (number of sensors) information. But AIBO Remote Framework works only for ERS-7. So in the next release, GetRobotDesign() returns DESIGN_ERS7 all the time. Regarding (3) In RemoteTest, the 3D button is disabled before getting sensor data, This is the intended behavior of RemoteTest. AIBO's 3D model is updated when RemoteTest gets sensor data from AIBO. When sensor data starts to be sent, RequestNetService( SENSOR_ID ) is called. This is the reason that GetRobotDesign() returns the correct AIBO model. The joint data of AIBO's 3D model is invalid before sensor data is sent. Because of this, the 3D button is disabled before starting to send sensor data. Thanks, AIBO SDE Support |
| V 1151 Say something progAibo [2004/06/09-23:44] |
| | Response : Edit : Delete |
|
| Hello, | i am using R-code, but how does Aibo (ERS-7) can say something ? i have | searched the order, but i have not found... | |
| +- 1154 RE:Say something AiboPet [2004/06/10-23:45] |
| | Response : Edit : Delete |
|
| >i
am using R-code, but how does Aibo (ERS-7) can say something ? | | R-CODE is designed to play back canned motions (MTN files) sounds (WAV files) | and others (LED and MID). It does not have any Text-To-Speech features. | | You must create the WAV file on your PC first in the right format. I typically | use 8bit 8kHz mono format (the ERS-7 supports other formats including 16kHz | ADPCM - good for voice). | Almost any WAVE editor will do. | | Then the WAV file must be combined to an action that can be played (WAV file | placed into AUDIO.ODA and ERS-7.MWC updated). | | The recently released RTool program will do this (read the instructions for | RTool if you are using that) | | However, I strongly suggest you start with YART_APG7. You supply the WAV files | and it does all this work for you, including automatically generating head- | moving and mouth-moving motions based on what is said in the WAV file. | http://www.aibohack.com/ers7/yart7.htm | [YART_APG7 can be used with or without YART7] | |
| |- 1165 RE:Say something progAibo [2004/06/14-21:37] |
| | Response : HomePage : Edit : Delete |
|
| Thanks ! | I'm going to read this. | |
| +- 1170 RE:Say something progAibo [2004/06/14-23:47] |
| | Response : HomePage : Edit : Delete |
|
|
i am using YART_APG, but i have an error message : "please turn off,
extra | information", to convert wav files, i am using the windows recorder by default. | (PCM 8khz, 8bit, mono). However what could i use ? Thanks again. | |
| +- 1173 RE:Say something progAibo [2004/06/15-18:28] |
| | Response : HomePage : Edit : Delete |
|
| oki,
i found an other soft, and it is very good ! | Aibo speaks ! :-) | the quality is poor ... too bad | |
| +- 1177 RE:Say something AiboPet [2004/06/15-23:55] |
| Response : Edit : Delete |
|
>the
quality is poor ... too bad Other audio suggestions: The recommended audio quality is 8kHz, 8 bit mono (works with all model AIBOs). For the ERS-7 you can use other rates (including 16kHz, 16 bit mono - which is overkill IMHO, and some inbetween). All the entertainment personalities on my site (from Cartman to Scooby Doo) use the 8kHz/8bit format. However, in my experience the biggest issue for playback is the quality of the original audio samples. Listen to the 8kHz/8bit version on your PC, and if it doesn't sound perfect (ie. no noise, no clipping, no artifacts other than the 8kHz sampling) then find another sample. Also be sure to normalize the volume of the audio sample (close to 90% full volume, but avoid clipping). Keep the audio at 16 bit resolution for all the editing, and do the final 8 bit conversion as the very last step. Also if the volume is too low, the mouth won't move far enough (when the mouth is closed it muffles the audio) If you have a choice of voice, pick something with a higher pitch (cartoon voices or female human voices usually work better). |
| V 1127 video streaming from Aibo to PC wangy22 [2004/06/03-04:57] |
| | Response : Edit : Delete |
|
| Hi, | | In past weeks I tried to implement the video streaming from Aibo to PC. The main | idea is to have a buffer (array of 44x38x3) in the Aibo storing the YUV data | received from Camera. Every time the main program is notified by FbkImageSensor, | it send the whole buffer to the client by TCP/IP connection. The client does the | YUV2RGB convertion. | | The problem is, the video received by the client is corrupted. Here are 3 | screenshots: | | http://msw.mcmaster.ca/~wangy/temp/s1.png | http://msw.mcmaster.ca/~wangy/temp/s2.png | http://msw.mcmaster.ca/~wangy/temp/s3.png | | However, if I make the server only transmit the first frame, the image received | by client is fine. | | The wireless mode for Aibo and PC is ad-hoc, so I think the network delay can be | ignored. | | Here is the code of Notify() to be called by subject FbkImageSensor: | | void gEye::Notify(const ONotifyEvent& event) | { | if (gEyeState == _IDLE) { | return; | //do nothing | } | | static int x, y; | | OFbkImageVectorData* fbkImageVectorData | = (OFbkImageVectorData*)event.Data(0); | | OFbkImageInfo* info = fbkImageVectorData->GetInfo(layer); | byte* data = fbkImageVectorData->GetData(layer); | | OFbkImage current_Y(info, data, ofbkimageBAND_Y); | OFbkImage current_Cr(info, data, ofbkimageBAND_Cr); | OFbkImage current_Cb(info, data, ofbkimageBAND_Cb); | | gEye_snd_data.width = info->width; | gEye_snd_data.height = info->height; | | if ((current_Y.IsValid() == true) | && (current_Cr.IsValid() == true) | && (current_Cb.IsValid() == true)){ | | for (y = 0; y < info->height - 1; y++){ | for (x = 0; x < info->width; x++){ | gEye_snd_data.image[x][y][0] = | current_Y.Pixel(x, y); | gEye_snd_data.image[x][y][1] = | current_Cr.Pixel(x, y); | gEye_snd_data.image[x][y][2] = | current_Cb.Pixel(x, y); | } | } | Snd(); | } | observer[event.ObsIndex()]->AssertReady(event.SenderID()); | } | | Any helps are appreciated. | | Yu | |
| |- 1128 RE:video streaming from Aibo to PC AiboPet [2004/06/03-23:40] |
| | | Response : Edit : Delete |
|
| | ... | | >The problem is, the video received by the client is corrupted... | | | | >However, if I make the server only transmit the first frame, the image | | received by client is fine. | | | | My guess is the old data has not being completely transmitted before the next | | Image notification comes in (and over-writes the old data). | | | | This will depend what is going on in your 'Snd()' routine. It should lock out | | the buffer access until all the data is transmitted, and/or copy it to the | | TCP/IP buffer first. | | In most of the other TCP/IP samples (eg: EchoServer/Client) copy the data to a | | transmit buffer before starting the send process ('SetSendData' or something | | like that, followed by 'Send'). The transmission is not complete until | | the 'SendCont' callback is called. | | |
| | |- 1131 RE:video streaming from Aibo to PC wangy22 [2004/06/04-03:14] |
| | | Response : Edit : Delete |
|
| | >... | | >>The problem is, the video received by the client is corrupted... | | > | | >>However, if I make the server only transmit the first frame, the image | | >received by client is fine. | | > | | >My guess is the old data has not being completely transmitted before the next | | >Image notification comes in (and over-writes the old data). | | | | That is exactly what I am thinking. Thanks for your confirm. =P | | | | > | | >This will depend what is going on in your 'Snd()' routine. It should lock out | | >the buffer access until all the data is transmitted, and/or copy it to the | | >TCP/IP buffer first. | | | | Here is my Snd() routine: | | | | OStatus gEye::Snd(){ | | if (Connection.state != _CONNECTED) | | return oFAIL; | | | | Connection.state = _SND; | | | | memcpy(Connection.snd_data, &gEye_snd_data, Connection.snd_size); | | TCPEndpointSendMsg sndMsg(Connection.endpoint, | | Connection.snd_data, | | Connection.snd_size); | | | | sndMsg.Call(IPStack, sizeof(TCPEndpointSendMsg)); | | | | if (sndMsg.error != TCP_SUCCESS) { | | OSYSLOG1((osyslogERROR, "%s : %s %d", | | "gEye::Snd()", | | "FAILED. sndMsg.error", sndMsg.error)); | | if (sndMsg.error == TCP_CONNECTION_CLOSED || | | sndMsg.error == TCP_CONNECTION_RESET || | | sndMsg.error == TCP_CONNECTION_TIMEOUT){ | | CloseConnection(); | | OpenConnection(); | | return oSUCCESS; | | }else{ | | Connection.state = _EXIT; | | return oFAIL; | | } | | } | | Connection.state = _CONNECTED; | | return oSUCCESS; | | } | | | | I didn't follow the way the samples did. | | | | >In most of the other TCP/IP samples (eg: EchoServer/Client) copy the data to a | | >transmit buffer before starting the send process ('SetSendData' or something | | >like that, followed by 'Send'). The transmission is not complete until | | >the 'SendCont' callback is called. | | | | What is that so? Is it because SendCont() is an extra entry point? | | | | Thanks a lot! | | |
| | +- 1134 RE:video streaming from Aibo to PC wangy22 [2004/06/04-04:16] |
| | | Response : Edit : Delete |
|
| | >... | | >>The problem is, the video received by the client is corrupted... | | > | | >>However, if I make the server only transmit the first frame, the image | | >received by client is fine. | | > | | >My guess is the old data has not being completely transmitted before the next | | >Image notification comes in (and over-writes the old data). | | > | | >This will depend what is going on in your 'Snd()' routine. It should lock out | | >the buffer access until all the data is transmitted, and/or copy it to the | | >TCP/IP buffer first. | | >In most of the other TCP/IP samples (eg: EchoServer/Client) copy the data to a | | >transmit buffer before starting the send process ('SetSendData' or something | | >like that, followed by 'Send'). The transmission is not complete until | | >the 'SendCont' callback is called. | | | | What happens after the following statement is executed? | | | | sendMsg.Send(ipstackRef, myOID_, Extra_Entry[entrySendCont], | | sizeof(TCPEndpointSendMsg)); | | | | Will it return immediately or wait until SendCont() has returned? | | |
| | +- 1137 RE:video streaming from Aibo to PC AiboPet [2004/06/04-22:21] |
| | | Response : Edit : Delete |
|
| | [two
replies merged] | | > Here is my Snd() routine:... | | > I didn't follow the way the samples did. | | | | > sndMsg.Call(IPStack, sizeof(TCPEndpointSendMsg)); | | | | That's one way to do it. It *should* block until the SendMsg is processed. | | That may cause problems since this can take a long time. | | | | However, in this case I'm not sure if all other notifications are blocked. | | I had some strange problems trying to use Call before (and went back to using | | Send and a second callback) | | | | ------ | | >What happens after the following statement is executed? | | >sendMsg.Send(ipstackRef, myOID_, Extra_Entry[entrySendCont], | | >sizeof(TCPEndpointSendMsg)); | | | | That's the other way - and the way I use all the time (also in the samples) | | and is recommended. | | The statement returns immediately. When the request is completed, it will call | | your routine "SendCont". | | | | In this case there are two radically different "Send"s | | The sendMsg structure contains the TCP/IP Send request (ie. send data from | | AIBO to PC). | | The "Send" function being called is the OPEN-R object communication notion of | | send (ie. send data from your process to the TCP/IP handler/stack) | | | | I suggest reworking the code to use this technique, and I suspect the | | collision problem will go away. | | |
| | |- 1140 RE:video streaming from Aibo to PC wangy22 [2004/06/08-04:42] |
| | | Response : Edit : Delete |
|
| | >[two
replies merged] | | >> Here is my Snd() routine:... | | >> I didn't follow the way the samples did. | | > | | >> sndMsg.Call(IPStack, sizeof(TCPEndpointSendMsg)); | | > | | >That's one way to do it. It *should* block until the SendMsg is processed. | | >That may cause problems since this can take a long time. | | > | | >However, in this case I'm not sure if all other notifications are blocked. | | >I had some strange problems trying to use Call before (and went back to using | | >Send and a second callback) | | > | | >------ | | >>What happens after the following statement is executed? | | >>sendMsg.Send(ipstackRef, myOID_, Extra_Entry[entrySendCont], | | >>sizeof(TCPEndpointSendMsg)); | | > | | >That's the other way - and the way I use all the time (also in the samples) | | >and is recommended. | | >The statement returns immediately. When the request is completed, it will call | | >your routine "SendCont". | | > | | >In this case there are two radically different "Send"s | | >The sendMsg structure contains the TCP/IP Send request (ie. send data from | | >AIBO to PC). | | >The "Send" function being called is the OPEN-R object communication notion of | | >send (ie. send data from your process to the TCP/IP handler/stack) | | > | | >I suggest reworking the code to use this technique, and I suspect the | | >collision problem will go away. | | | | OK, I've re-written the code by using the suggested technique. Unfortunately, | | the problem still exists. | | | | Then I tried to make the program wait for a key strike between each frame is | | recieved, by adding getchar(), so that the buffer is not over written by the | | next frame while the TCP/IP packet is being sent. But the image it got is still | | gappy. | | | | Any ideas? | | | | Thanks a lot! | | |
| | +- 1142 RE:video streaming from Aibo to PC wangy22 [2004/06/08-07:43] |
| | Response : Edit : Delete |
|
| OK,
I just found the problem. The server program in Aibo is OK. But in the | client program, I made GTK drawing the image from the buffer at the same time of | reading the date from TCP/IP into the buffer. | | Thanks a lot for your help! | |
| +- 1176 RE:video streaming from Aibo to PC jcbaillie [2004/06/15-21:54] |
| Response : HomePage : Edit : Delete |
|
Hi, Check this, you will find video streaming among other things: http://urbi.sourceforge.net JCB >Hi, > >In past weeks I tried to implement the video streaming from Aibo to PC. The main >idea is to have a buffer (array of 44x38x3) in the Aibo storing the YUV data >received from Camera. Every time the main program is notified by FbkImageSensor, >it send the whole buffer to the client by TCP/IP connection. The client does the >YUV2RGB convertion. > >The problem is, the video received by the client is corrupted. Here are 3 >screenshots: > >http://msw.mcmaster.ca/~wangy/temp/s1.png >http://msw.mcmaster.ca/~wangy/temp/s2.png >http://msw.mcmaster.ca/~wangy/temp/s3.png > >However, if I make the server only transmit the first frame, the image received >by client is fine. > >The wireless mode for Aibo and PC is ad-hoc, so I think the network delay can be >ignored. > >Here is the code of Notify() to be called by subject FbkImageSensor: > >void gEye::Notify(const ONotifyEvent& event) >{ > if (gEyeState == _IDLE) { > return; > //do nothing > } > > static int x, y; > > OFbkImageVectorData* fbkImageVectorData > = (OFbkImageVectorData*)event.Data(0); > > OFbkImageInfo* info = fbkImageVectorData->GetInfo(layer); > byte* data = fbkImageVectorData->GetData(layer); > > OFbkImage current_Y(info, data, ofbkimageBAND_Y); > OFbkImage current_Cr(info, data, ofbkimageBAND_Cr); > OFbkImage current_Cb(info, data, ofbkimageBAND_Cb); > > gEye_snd_data.width = info->width; > gEye_snd_data.height = info->height; > > if ((current_Y.IsValid() == true) > && (current_Cr.IsValid() == true) > && (current_Cb.IsValid() == true)){ > > for (y = 0; y < info->height - 1; y++){ > for (x = 0; x < info->width; x++){ > gEye_snd_data.image[x][y][0] = >current_Y.Pixel(x, y); > gEye_snd_data.image[x][y][1] = >current_Cr.Pixel(x, y); > gEye_snd_data.image[x][y][2] = >current_Cb.Pixel(x, y); > } > } > Snd(); > } > observer[event.ObsIndex()]->AssertReady(event.SenderID()); >} > >Any helps are appreciated. > >Yu |
| V 1125 wireless control of Aibo djtommyc [2004/06/02-01:13] |
| | Response : Email : Edit : Delete |
|
| I
just started working with the aibo and was wondering what software I
would | need to do the following. I want to be able to give the Aibo wireless commands. | something where I could type in commands in a console on a machine and give the | aibo commands. | | thanks for the help, | tom | |
| |- 1126 RE:wireless control of Aibo AiboPet [2004/06/02-23:50] |
| | | Response : Edit : Delete |
|
| | >I
just started working with the aibo and was wondering what software I
would | | >need to do the following. I want to be able to give the Aibo wireless | | commands. | | > something where I could type in commands in a console on a machine and give | | the | | >aibo commands. | | | | There are several options. | | The easiest is to use RCODE and the telemetry pipe. I do it all the time, | | [there are OpenR SDK samples that provide telnet-like commands, but RCODE has | | much much more built-in technology] | | ---- | | Download the RCODE SDK from this Website [ERS-7]. RCODE for the ERS-210/220 is | | found elsewhere [the same telnet approach works for any WiFi enabled RCODE] | | Install the RCODE stick image on an 8 or 16 MB Programmable Memory Stick (be | | sure it doesn't have an ERF- number) | | Configure the WLAN connection (by editing WLANCONF.TXT) | | Boot up AIBO. Be sure you can ping your AIBO from the PC (eg: 'ping | | 10.0.1.100') | | | | From your PC, connect to the AIBO using 'telnet' to port 21002 | | [eg: 'telnet 10.0.1.100 21002' - assuming you use the default IP address] | | Then you will see some information displayed, and you can type in RCODE | | commands directly [no line editing]. For example: | | PLAY ACTION SIT | | PLAY ACTION WALK 0 1000 | | PRINT "%d" Batt_Rest | | | | (etc). | | There are many sophisticated ways of using this feature, and a number of tools | | that exploit it (including AibNet and most of my AiboRemote-like programs) | | |
| | +- 1160 RE:wireless control of Aibo djtommyc [2004/06/12-03:03] |
| | | Response : Email : Edit : Delete |
|
| | >>I
just started working with the aibo and was wondering what software I
would | | >>need to do the following. I want to be able to give the Aibo wireless | | >commands. | | >> something where I could type in commands in a console on a machine and give | | >the | | >>aibo commands. | | > | | >There are several options. | | >The easiest is to use RCODE and the telemetry pipe. I do it all the time, | | >[there are OpenR SDK samples that provide telnet-like commands, but RCODE has | | >much much more built-in technology] | | >---- | | >Download the RCODE SDK from this Website [ERS-7]. RCODE for the ERS-210/220 is | | >found elsewhere [the same telnet approach works for any WiFi enabled RCODE] | | >Install the RCODE stick image on an 8 or 16 MB Programmable Memory Stick (be | | >sure it doesn't have an ERF- number) | | >Configure the WLAN connection (by editing WLANCONF.TXT) | | >Boot up AIBO. Be sure you can ping your AIBO from the PC (eg: 'ping | | >10.0.1.100') | | > | | >From your PC, connect to the AIBO using 'telnet' to port 21002 | | >[eg: 'telnet 10.0.1.100 21002' - assuming you use the default IP address] | | >Then you will see some information displayed, and you can type in RCODE | | >commands directly [no line editing]. For example: | | >PLAY ACTION SIT | | >PLAY ACTION WALK 0 1000 | | >PRINT "%d" Batt_Rest | | > | | >(etc). | | >There are many sophisticated ways of using this feature, and a number of tools | | >that exploit it (including AibNet and most of my AiboRemote-like programs) | | | | I'm useing an ers210 would these instructions be any differnt in that case? I | | couldn't find a file named "WLANCONF.TXT" but there was one named "WLANDFLT.TXT". | | |
| | +- 1161 RE:wireless control of Aibo AiboPet [2004/06/12-23:38] |
| | Response : Edit : Delete |
|
| >>>..
I want to be able to give the Aibo wireless commands | ... | >>The easiest is to use RCODE and the telemetry pipe. | ^^ typo: I call the 21002 the RCODE pipe (the telemetry pipe is something else) | | UPDATE: | also see recent FAQ updates: | https://openr.aibo.com/openr/eng/perm/rcode/regi_faq3.php4#B0005 | https://openr.aibo.com/openr/eng/perm/rcode/regi_faq3.php4#C0001 | | ------ | >I'm useing an ers210 would these instructions be any differnt in that case? I | >couldn't find a file named "WLANCONF.TXT" but there was one | named "WLANDFLT.TXT". | | The ERS-210/220 version of RCODE is very similar. | The telnet connection is the same (port 21002) | More info for RCODE for the 210/220: http://www.us.aibo.com/openr_code.htm | | ----- | HOWEVER you found the one very important difference. | The location of the WLANCONF.TXT file is different. | | ERS-7 RCODE (and other newer software like AiboMind, AiboEyes(210) and OpenR | SDK) | \OPEN-R\SYSTEM\CONF\WLANCONF.TXT | | ERS-2x0 RCODE (and older software like Navigator and Painter) | OPEN-R\SYSTEM\CONF\WLAN\HTML\WLANCONF.TXT | | Copy the WLANDFLT.TXT file from that folder to WLANCONF.TXT | (and edit the WLANCONF.TXT file if necessary). | The WLANDFLT.TXT is the default setting and can be used for starting values. | In newer software look for the file "SAMPLE\WLANCONF.TXT". | |
| +- 1175 RE:wireless control of Aibo jcbaillie [2004/06/15-21:53] |
| Response : HomePage : Edit : Delete |
|
Hi, check this: http://urbi.sourceforge.net JCB >I just started working with the aibo and was wondering what software I would >need to do the following. I want to be able to give the Aibo wireless commands. > something where I could type in commands in a console on a machine and give the >aibo commands. > >thanks for the help, >tom |
| V 1163 Sound programming in AIBO wangy22 [2004/06/13-06:23] |
| | Response : Edit : Delete |
|
| Hi, | | I am currently implementing audio streaming from Aibo to PC, and I have several | questions here: | | 1. what do oprmreqMIC_OMNI and oprmreqMIC_UNI mean? Do they stand for stereo and | mono? | | 2. What is the time length of the audio data for each OSoundVectorData when | Notify() routine is called? Is it fixed to 32ms? | | 3. What is the largest size for setting each TCP/IP packet? | | 4. What does ALC do? | | Thanks a lot! | |
| |- 1169 RE:Sound programming in AIBO AiboPet [2004/06/14-23:37] |
| | Response : Edit : Delete |
|
| >I
am currently implementing audio streaming from Aibo to PC, and I have | several | | BTW: if you have an ERS-7, then I suggest you look into the Remote Framework. | It has audio-streaming built-in (in both directions) -- but a different set of | features and trade-offs. | | >1. what do oprmreqMIC_OMNI and oprmreqMIC_UNI mean? Do they stand for stereo | and | >mono? | | OMNI-Directional and UNI-Directional. | See Udi's page for comparing the two modes. | http://www.benreuven.com/micresponse.htm | [done on the ERS-210] | | >2. What is the time length of the audio data for each OSoundVectorData when | >Notify() routine is called? Is it fixed to 32ms? | | The size of the data is fixed, at 512 samples (4 bytes per sample, 512/16000 = | 32ms). | IIRC the ERS-7 version is 256 samples. | | >3. What is the largest size for setting each TCP/IP packet? | I believe it can handle larger sizes, but I try to keep it under 32000 bytes | (for a number of reasons). | | >4. What does ALC do? | Automatic Level Control. | AFAIK it doesn't change much | |
| +- 1174 RE:Sound programming in AIBO jcbaillie [2004/06/15-21:52] |
| Response : HomePage : Edit : Delete |
|
The
solution of all your problems is here: http://urbi.sourceforge.net JCB >Hi, > >I am currently implementing audio streaming from Aibo to PC, and I have several >questions here: > >1. what do oprmreqMIC_OMNI and oprmreqMIC_UNI mean? Do they stand for stereo and >mono? > >2. What is the time length of the audio data for each OSoundVectorData when >Notify() routine is called? Is it fixed to 32ms? > >3. What is the largest size for setting each TCP/IP packet? > >4. What does ALC do? > >Thanks a lot! |
| V 1172 problem with 16meg stick Mario Alleyne [2004/06/15-15:21] |
| Response : Email : Edit : Delete |
|
I
did something stupid. I have an ERS 210A and I put my 16 meg pink memory stick into a non Sony memory stick reader to change programs. It worked several times but now Aibo will not boot up with this stick any more. He works fine with my other sticks but this one seems to be toast. Is there any way to fix this 16 meg stick? I live in Canada and they are soo very hard to get.HELP my little doggy please. |
| V 1158 ERS 220 Rcode djtommyc [2004/06/11-02:53] |
| | Response : Email : Edit : Delete |
|
| I've
been trying to work with rcode on the ers220. Most of the faq on this
site | are about the ers 7. Is there a place where I can find lots of info about rcode | on the ers 220? | | thanks | |
| +- 1162 RE:ERS 220 Rcode AiboPet [2004/06/12-23:40] |
| Response : Edit : Delete |
|
>
Is there a place where I can find lots of info about rcode >on the ers 220? > More info for RCODE for the 210/220/31x: http://www.us.aibo.com/openr_code.htm Reference manual: http://www.us.aibo.com/rcode/rcode12ref.txt RCodePlus (my addon): http://www.aibohack.com/rcode |
| V 1145 FAQs for the AIBO Remote Framework have been added openrsupport@(Administrator) [2004/06/09-00:47] |
| | Response : Edit : Delete |
|
| FAQs for the AIBO Remote Framework have been added. | | Setup, Technical, and Trouble shooting FAQs have been added. | Please read them when you want to know how to use the AIBO | Remote Framework or you meet some problem using the AIBO | Remote Framework. | | Thank you, | | AIBO SDE Support | |
| +- 1150 RE:FAQs for the AIBO Remote Framework have been added phebas [2004/06/09-19:19] |
| | Response : Email : Edit : Delete |
|
| These
do not print properly, the text runs of the left of the page | | phebas | | >FAQs for the AIBO Remote Framework have been added. | > | >Setup, Technical, and Trouble shooting FAQs have been added. | >Please read them when you want to know how to use the AIBO | >Remote Framework or you meet some problem using the AIBO | >Remote Framework. | > | >Thank you, | > | >AIBO SDE Support | |
| +- 1157 RE:FAQs for the AIBO Remote Framework have been added openrsupport@(Administrator) [2004/06/11-02:02] |
| Response : Edit : Delete |
|
There
are several ways to print this out. 1 - Set your paper orientation to landscape mode. 2 - Use a reduction feature (zoom) that your printer has. 3 - Copy the text to an editor and configure the printout through your editor. We apologize that the FAQ is not directly printable. This will be fixed when the overall website templates have been changed, but we currently cannot say when this will happen. Best regards, AIBO SDE Support |
| V 1155 ERS-7 charging station strange behavior outsider [2004/06/11-00:01] |
| Response : Edit : Delete |
|
Hi every one I have an ERS-7 model. The other day happenend something strange with the charger. It suddenly rejected to charge my robot dog. When the charger has no Aibo on it, nothing appears on the LCD screen, unless I press a button. When I press a button, the LCD shows for some seconds the 'Battery full' graphic in a flashing manner and then nothing. If I connect my dog to it, the dog detects that is attached to the charger since the 'Pause button' appears in red, but neither charge transfer occurs nor anything is displayed on the LCD screen. If I press any charger button, then the same graphic as before appears. What does it means?. How can I repair it?. Should I bring it to some Sony special place?. It is still under warranty period... but I would like to avoid it since I suspect that it will take a long time and I have to finish some experiments. The dog is working properly, until batteries are gone... Please, any advice will be wellcome. |
| V 1146 AIBO Remote Framework has been released openrsupport@(Administrator) [2004/06/09-00:48] |
| | Response : Edit : Delete |
|
| AIBO Remote Framework has been released. | You can get these files from the download page on this site. | | Thank you, | | AIBO SDE Support | |
| +- 1147 RE:AIBO Remote Framework has been released openrsupport@(Administrator) [2004/06/09-02:01] |
| | Response : Edit : Delete |
|
| There
are a few errors in the download page which prevents you from
downloading | all of the necessary files for the AIBO Remote Framework. We will notify you | when the download page has been fixed. We apologize for this inconvenience. | | AIBO SDE Support | |
| +- 1152 RE:AIBO Remote Framework has been released openrsupport@(Administrator) [2004/06/10-00:21] |
| Response : Edit : Delete |
|
The
download page has been fixed. All AIBO Remote Framework downloads are
now available. We apologize for the delay. AIBO SDE Support |
| V 1148 AIBO shut down with Remote Framework mbarnig [2004/06/09-05:21] |
| | Response : Email : Edit : Delete |
|
| I
installed the VAIBOserver for ERS-7, put the Remote Framework sample | executables in the same folder, copied the RF Memory Stick Image to a 16 Mbyte | AIBO Memory Stick, customized the WLANCONF.txt file and booted the AIBO with | the new Memory Stick. The server and the excutables are working on the PC, but | the AIBO shut down after booting, before the WLAN LED lights up. Do I need to | copy other files on the Memory Stick ? | Thank you for your help | Marco Barnig, Luxembourg | |
| +- 1149 RE:AIBO shut down with Remote Framework mbarnig [2004/06/09-05:46] |
| Response : Email : Edit : Delete |
|
Now
it works fine. I think it was only a battery problem. The WLAN LED does not light. Is this normal ? greetings Marco |
| V 1143 FAQ B0005 for the R-CODE SDK has been added. openrsupport@(Administrator) [2004/06/09-00:41] |
| Response : Edit : Delete |
|
FAQ
B0005 for the R-CODE SDK has been added. This FAQ answers the following: Q. How do I setup the wireless LAN console for the R-CODE SDK? AIBO SDE Support |
| V 1139 protected mode (Administrators) GriffithJF [2004/06/07-12:04] |
| | Response : Edit : Delete |
|
| I
just noticed a reply to another post by open-r support team, so I
though I'd ask: | | Is there a way to use the TCPGateway in memory protected mode yet? | If not are there any plans to add this capability? | | Thanks | |
| +- 1141 RE:protected mode (Administrators) openrsupport@(Administrator) [2004/06/08-06:42] |
| Response : Edit : Delete |
|
>Is
there a way to use the TCPGateway in memory protected mode yet? No. >If not are there any plans to add this capability? No, there is no plan. But, on the PC side, you can use other, more powerful memory debuging tools; "electric fence" and so on. Thanks, AIBO SDE Support |
| V 1133 Wireless LAN Card for Aibo ers210 l337Nick [2004/06/04-04:05] |
| | Response : Edit : Delete |
|
| We
are thinking of using remote processing with Aibo and we would like to
know | if any IEEE802.11b/IEEE802.11 standard PCMCIA WLAN card would work with Aibo. | Thankyou | |
| +- 1136 RE:Wireless LAN Card for Aibo ers210 openrsupport@(Administrator) [2004/06/04-11:44] |
| | Response : Edit : Delete |
|
| Dear l337Nick, | | Do you mean PCMCIA WLAN card to be inserted into AIBO [ERS-210 or ERS-220]? Or | do you mean PCMCIA WLAN card for your PC? | | For AIBO, ERS-7 does not require a WLAN card to be inserted into the robot, as | ERS-7 comes equipped with wireless capabilities. ERS-210/220 does require the | ERA-201D1 PCMCIA WLAN card to be inserted into your robot. | | For your PC, any IEEE802.11b compliant card should work. | | If you want to use AIBO ERS-7 with AIBO Remote Framework, | please see AIBO Remote Framework FAQ A0004 | https://openr.aibo.com/openr/eng/perm/rfw/regi_faq2.php4#A0004 | | Thanks, | | AIBO SDE Support | |
| +- 1138 RE:Wireless LAN Card for Aibo ers210 l337Nick [2004/06/05-01:04] |
| Response : Edit : Delete |
|
>Dear
lNick, > >Do you mean PCMCIA WLAN card to be inserted into AIBO [ERS or ERS]? Or >do you mean PCMCIA WLAN card for your PC? > >For AIBO, ERS does not require a WLAN card to be inserted into the robot, as >ERS comes equipped with wireless capabilities ERS does require the >ERAD PCMCIA WLAN card to be inserted into your robot > >For your PC, any IEEEb compliant card should work > >If you want to use AIBO ERS with AIBO Remote Framework, >please see AIBO Remote Framework FAQ A >httpsopenraibocomopenrengpermrfwregi_faqphp#A > >Thanks, > >AIBO SDE Support So any IEEE802.11b OR IEEE802.11g WLAN PCMCIA card will work with the ERS ? Thanks for the reply (The article that 2004/06/05-01:02 posted, was modified by Author) |
| V 1132 AIBO Remote Framework l337Nick [2004/06/04-03:55] |
| | Response : Edit : Delete |
|
| When will the AIBO Remote Framework be released? | |
| +- 1135 RE:AIBO Remote Framework openrsupport@(Administrator) [2004/06/04-11:41] |
| Response : Edit : Delete |
|
Dear l337Nick, Thank you for your message. AIBO Remote Framework will be released soon. Please check back at this website for further details. Please see AIBO Remote Framework FAQ A0016 https://openr.aibo.com/openr/eng/perm/rfw/regi_faq2.php4#A0016 Thank you, AIBO SDE Support (The article that 2004/06/04-11:37 posted, was modified by Author) |
| V 1052 remote processing bug ckch [2004/04/19-11:22] |
| | Response : Edit : Delete |
|
| In
the open-r 1.1.5, the remote processing cannot broadcast messages to | different objects. That is, for example, I cannot do this in the HOSTGW.CFG: | | TCPGateway.In1.S TCPGateway.Out1.O | TCPGateway.In1.S TCPGateway.Out2.O | TCPGateway.In1.S TCPGateway.Out3.O | | Other robotcuppers have the same problem as well, and now in the competiton, | it causes a lot of problems. | | Could the SDK people please find out a solution for this? This problem doesnt | exist in the old openr. | | Thanks a lot :-} | | ckch | |
| +- 1130 RE:remote processing bug openrsupport@(Administrator) [2004/06/04-02:40] |
| Response : Edit : Delete |
|
Please
refer to the new ObjectComm-multi sample programs. They demonstrate how to have multiple subject-observer connections using Remote Processing OPEN-R. Thank you. OPEN-R SDE Support |
| V 1129 ObjectComm-multi sample programs have been released openrsupport@(Administrator) [2004/06/04-02:32] |
| Response : Edit : Delete |
|
ObjectComm-multi
sample programs have been released. These samples explain how to use multi-object communication with Remote Processing OPEN-R. Thank you. OPEN-R SDE Support |
| V 1123 Installation and Compiling l337Nick [2004/06/01-22:45] |
| | Response : Edit : Delete |
|
| I am
running Fedora Core (Linux) and none of the sample programs will
compile, | Open-R SDK will not install properly. None of the developer tools will install. | Everything brings up errors. | Can anyone help? | |
| +- 1124 RE:Installation and Compiling l337Nick [2004/06/01-22:59] |
| Response : Edit : Delete |
|
>I
am running Fedora Core (Linux) and none of the sample programs will
compile, >Open-R SDK will not install properly. None of the developer tools will install. >Everything brings up errors. However when we use the cygwin packages in windows it seems to work Why is this? |
| V 1122 how to send commands to the legs and heads simultaneously totoduke [2004/06/01-01:55] |
| Response : Email : Edit : Delete |
|
Hi, Does anybody know how to send commands to the legs and head simultaneously? I am now doing the researches on ERS7. I need to control the legs and head at the same time. What I did is as following: 1) define the joint locator (including leg joints and head joints) in the head file. 2) get the primitive ID in the initialization step 3) set the joint Gains to the legs 4) calibrating the joint values of the legs 5) set the joint values to the legs and heads. Here I did not set the joint gains and the calibrate the joint vavlues to the head joints. When I debug, I found the jointID of all joints are right. But it seemed that the values of joints I wanted to send to head joints were always sent to the leg joints. By the way, the control of the leg joints is perfect. Did anyone encounter such a situation before? Is there anything wrong with what i did? Thanks a lot. |
| V 1105 Make aibo do the difference between 2 balls angers2004 [2004/05/25-22:47] |
| | Response : Edit : Delete |
|
| Morning, | | I'm working (for research purposes) on an AIBO ERS-220. I would like to | implement a basic object recognition based on colors. For example, if Aibo see | its pink ball, it dances, else if the ball is blue, it sits. Is it possible? | With R-code or Open-R ? Which is the best way to understand how Aibo can do | that? | | Thanks for your help, regards, | | Thibaud | |
| +- 1106 RE:Make aibo do the difference between 2 balls AiboPet [2004/05/25-23:41] |
| | Response : Edit : Delete |
|
| >I'm
working (for research purposes) on an AIBO ERS-220. I would like to | >implement a basic object recognition based on colors. For example, if Aibo see | >its pink ball, it dances, else if the ball is blue, it sits. Is it possible? | >With R-code or Open-R ? Which is the best way to understand how Aibo can do | >that? | | Yes, it is relatively easy in either the OpenR SDK or R-CODE (RCodePlus for | the 220 actually). In the very simple case (ie. a single reaction to a color | regardless of location), this can be done immediately with YART [ie. drag and | drop, no real coding required] | ---- | In the OpenR SDK case, look at the ImageObserver or BallTrackingHead samples. | Both of these setup the CDT color detection thresholds to detect the pink ball. | See "SetCdtVectorOfPinkBall". | This can be extended to add other color ranges. Each color range turns into a | single bit in the 88x77 pixel/byte CDT image. | The hardware will give you a rough count of how many pixels fall in that color | range ('ColorFrequency', very fast). The ball tracking logic also shows how | you can calculate the centroid of the color image and can be turned into a | horizontal/vertical angle from the nose. | | ---- | RCODE and RCodePlus provide support for the "Pink_Ball" variables. | With RCodePlus (210/220/31x), there are RCODE extensions that provide | functionality for other colors: | http://www.aibohack.com/rcode/rcp_image.htm | | The default color ranges are used (but can be tweeked if you don't like them). | Ball color is pink/red. The "favorite" color is blue. The "unfavorite" color | is yellow/green. | | AP_COLORLVL returns the (very fast) ColorFrequency exactly the same as the | OpenR SDK. | AP_COLORFND calculates a centroid in a very similar manner to the | BallTrackingHead sample | | If you look at the default R-CODE.R that comes with RCodePlus, it has a | trivial example of color detection [YART compatible] | |
| +- 1112 RE:Make aibo do the difference between 2 balls angers2004 [2004/05/27-00:13] |
| | Response : Edit : Delete |
|
| (Thanks
a lot AIBOpet for the information that you gave us.) | | I would like to know if it were possible in R-Code or Open-R to make AIBO | recognise different shapes and not only geometrical shapes, a shape like a | hand or something else with the ERS-220. | |
| +- 1120 RE:Make aibo do the difference between 2 balls AiboPet [2004/05/30-02:03] |
| Response : Edit : Delete |
|
>I
would like to know if it were possible in R-Code or Open-R to make AIBO >recognise different shapes and not only geometrical shapes, a shape like a >hand or something else with the ERS-220. With Open-R you get the raw image data (perhaps using the smaller 88x72 pixel image or 88x72 color map) You can perform your own shape recognition algorithm on it. Written in C++, you can make if relatively fast depending on the algorithm. ---- In RCodePlus, the AP_COLORFND primitive doesn't give much information. Processing each pixel (88x72 = 6336 "pixels") would be too slow in RCODE There is a slightly more complicated AP_COLORGRID that reduces the 88x72 color map image to something smaller. With the reduced image (say 8x7 cells) it should be small enough to run in RCODE. In either case, you have to invent the shape detection algorithm yourself. BTW: hand detection is mostly triggered by color (Flesh color in TR.CT) ------ BTW: The ERS-7 has the Evolution Robotics object recognition built in to the AiboMind software. However it is proprietary. Other alternatives: If you can rely on a WiFi connection, I have used the free-ware OpenCV image library from Intel. It has similar features. It has not been ported to the OpenR SDK, so it must run on a PC. [image captured on AIBO, transmitted and processed on PC - AIBO can react to the results] If you are looking at more serious algorithms running on the dog itself, you will need to use the OpenR SDK Take a look at the various RoboCup sources, they have invented their own - but primarily for detecting objects on the soccer field and mapping it to a world view. The CMU Vision module for example: http://www-2.cs.cmu.edu/~jbruce/cmvision is the one used by the Tekkotsu framework |
| V 1118 Motion Editor under Linux outsider [2004/05/29-01:50] |
| Response : Edit : Delete |
|
Hi,
I would like to say that I have tested Motion editor under Linux using
Wine, and it works perfectly ( My system is a Pentium III 1Ghz with 256 Mb). I have been able to run the program, create some movements and save the motion file. No problems found during my execution. Since my working environment is Linux, I will use the ME all the time on that system and report any bug that I find. To run ME with wine, just intall wine (http://www.winehq.com) and then type 'wine MEdit.exe' |
| V 1101 How to control Aibo ERS-7 face LEDs layer? outsider [2004/05/24-16:08] |
| | Response : Edit : Delete |
|
| Hi, | I'm trying to control ERS-7 face LEDs layers. Even that I do specify the 'mode' | in the OLEDCommandValue3 variable, Aibo doesn't pay attention to it. So I can | only access one of the layers (either A or B), but I cannot change between layer | inside a program by changing the 'mode' value. | How can I access the other layer?. How can I dinamically change between layers | during the execution of a program?. | | Thank you very much. | |
| +- 1107 RE:How to control Aibo ERS-7 face LEDs layer? GriffithJF [2004/05/26-14:44] |
| | Response : Email : Edit : Delete |
|
| The
whole led bank can only be in one mode at a time. Even though all the
leds | have a mode only the mode of the highest led you set will take affect. | | eg: If in your command to OVirtual Robot you send | led0, modeA | led1, modeB | Then the whole bank will be in modeB | |
| +- 1115 Your explanation is does not work outsider [2004/05/28-02:22] |
| | Response : Edit : Delete |
|
| Thank
you for your answer to my question. | Nevertheless your explanation does not work. I have tested it, having several | lights in mode A and several in mode B. The system remains in the first mode | selected and there is no way to change to another mode. | If I have the program running and I put some LEDs in one mode and the rest in | the other mode the mode that prevaledes is the first one. If I don't stop the | program and I continue putting all LEDs in the mode that has not prevaled, the | system does not allow to change to the other mode. | | The problem is that I cannot switch between modes. The question is, how can I | switch the face layers? | | Thank you | |
| +- 1116 RE:Your explanation is does not work GriffithJF [2004/05/28-09:44] |
| | Response : Email : Edit : Delete |
|
| Ok
I said the "highest LED". In my example its I believe LED1 would win not | because it comes later but because it is a higher number. Note: In our | experience this applies if you say anything at all about the led. Even if you | are turning it off. You cannot have leds in both modes at the same time. Try this: | | Set led14 to modeA (doesn't have to be on). Set the other leds to whatever modes | you like - modeA will win. | Set led14 to modeB. Set the other leds to whatever modes you like - modeB will win. | | It works for us. Now we may have just fluked it I admit :-) | |
| +- 1117 RE:Your explanation is does not work outsider [2004/05/28-17:12] |
| Response : Edit : Delete |
|
Thank
you very much for your answer and clear explanation. You are right, it works as you said and I missunderstood your first explanation. Thanks a lot, since the problem is solved. |
| V 1108 Forbidden Stack Dump of mystery GriffithJF [2004/05/26-14:49] |
| | Response : Email : Edit : Delete |
|
| Hi all, | I'm posting this far and wide to get an answer. Consider the following stack | dump from EMON.LOG | | 225cf110: 00000001 225cf808 ffffffff ffffffff ......\"........ | 225cf120: ffffffff ffffffff ffffffff ffffffff ................ | * | 225cf370: ffffffff 00000070 00000070 00000070 ....p...p...p... | 225cf380: 00000071 00000071 00000071 00000071 q...q...q...q... | | and a bit further on | 225cf5a0: 0000001c 0000001c 0000001c 0000001c ................ | * | 225cf690: 0000001d 0000001d 0000001d 0000001d ................ | * | 225cf6b0: 00000026 00000029 00000029 0000002b &...)...)...+... | 225cf6c0: 0000002b 0000002b 0000002e ffffffff +...+........... | 225cf6d0: ffffffff ffffffff ffffffff ffffffff ................ | * | 225cf790: ffffffff ffffffff 229d2e10 229e2dac ...........".-." | | We noticed this last year but never had enough test cases to investigate properly. | Why are rows missing from the stack dump (replaced with a lone *). | | Stack traces are difficult to interpret if the return address is in the | "forbidden zone". | | Anyone with creative theories, or even the actual answer? | |
| +- 1111 RE:Forbidden Stack Dump of mystery grbolar [2004/05/26-19:52] |
| | Response : Edit : Delete |
|
| Do
you prefer a verbose output like this ? | | 225cf110: 00000001 225cf808 ffffffff ffffffff ......\"........ | 225cf120: ffffffff ffffffff ffffffff ffffffff ................ | 225cf130: ffffffff ffffffff ffffffff ffffffff ................ | 225cf140: ffffffff ffffffff ffffffff ffffffff ................ | 225cf150: ffffffff ffffffff ffffffff ffffffff ................ | 225cf160: ffffffff ffffffff ffffffff ffffffff ................ | 225cf170: ffffffff ffffffff ffffffff ffffffff ................ | 225cf180: ffffffff ffffffff ffffffff ffffffff ................ | 225cf190: ffffffff ffffffff ffffffff ffffffff ................ | 225cf1a0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf1b0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf1c0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf1d0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf1e0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf1f0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf200: ffffffff ffffffff ffffffff ffffffff ................ | 225cf210: ffffffff ffffffff ffffffff ffffffff ................ | 225cf220: ffffffff ffffffff ffffffff ffffffff ................ | 225cf230: ffffffff ffffffff ffffffff ffffffff ................ | 225cf240: ffffffff ffffffff ffffffff ffffffff ................ | 225cf250: ffffffff ffffffff ffffffff ffffffff ................ | 225cf260: ffffffff ffffffff ffffffff ffffffff ................ | 225cf270: ffffffff ffffffff ffffffff ffffffff ................ | 225cf280: ffffffff ffffffff ffffffff ffffffff ................ | 225cf290: ffffffff ffffffff ffffffff ffffffff ................ | 225cf2a0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf2b0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf2c0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf2d0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf2e0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf2f0: ffffffff ffffffff ffffffff ffffffff ................ | 225cf300: ffffffff ffffffff ffffffff ffffffff ................ | 225cf310: ffffffff ffffffff ffffffff ffffffff ................ | 225cf320: ffffffff ffffffff ffffffff ffffffff ................ | 225cf330: ffffffff ffffffff ffffffff ffffffff ................ | 225cf340: ffffffff ffffffff ffffffff ffffffff ................ | 225cf350: ffffffff ffffffff ffffffff ffffffff ................ | 225cf360: ffffffff ffffffff ffffffff ffffffff ................ | 225cf370: ffffffff 00000070 00000070 00000070 ....p...p...p... | 225cf380: 00000071 00000071 00000071 00000071 q...q...q...q... | |
| +- 1113 RE:Forbidden Stack Dump of mystery GriffithJF [2004/05/27-10:21] |
| | Response : Email : Edit : Delete |
|
| >Do
you prefer a verbose output like this ? | > | >225cf110: 00000001 225cf808 ffffffff ffffffff ......\"........ | <snip> | | Frankly yes. Are you claiming that * just represents repeated lines? | Any proof? (Very interested) | |
| +- 1114 RE:Forbidden Stack Dump of mystery grbolar [2004/05/27-12:45] |
| Response : Edit : Delete |
|
>>Do
you prefer a verbose output like this ? >> >>225cf110: 00000001 225cf808 ffffffff ffffffff ......\"........ ><snip> > >Frankly yes. Are you claiming that * just represents repeated lines? >Any proof? (Very interested) No proof. But, it is the convention of the 'od' command. |
| V 1007 Dog crashed in ovirtualRobotComm kcph007 [2004/03/05-15:46] |
| | Response : Edit : Delete |
|
| Hi, | we're having a problem with one dog. It crashed and when I parsed the emon.log, | it said | context: 80234740 | object: ovirtualRobotComm | | The mem stick is working on other dog. | Is it problem of hardware? what that might be? How do I fix? | | Thanks alot. | |
| +- 1110 RE:Dog crashed in ovirtualRobotComm GriffithJF [2004/05/26-15:15] |
| Response : Email : Edit : Delete |
|
>Hi, >we're having a problem with one dog. It crashed and when I parsed the emon.log, >it said >context: 80234740 >object: ovirtualRobotComm > >The mem stick is working on other dog. >Is it problem of hardware? what that might be? How do I fix? Are you running protected memory? If not it is possible you have some bad pointer arithmetic or something that is causing you to write into the memory occupied by ovirtualRobotComm. If this happens all sorts of bad and unexplainable things can happen. The only drawback with protected meory is that the TCPGateway object doesn't work. But If you don't need it for testing give it a shot. |
| V 1058 TinyFTPD gijoe [2004/04/22-16:10] |
| | Response : HomePage : Edit : Delete |
|
| We've
been experiencing problems using TinyFTPD on our ERS-7s (haven't tested | ERS-210s). Basically, it's sometimes very very slow to upload (and sometimes | doesn't work at all), even for small files (~1kb or less). Other times it fails | to upload the entire file, particularly for larger files (~100kb). | | We didn't have this problem with our ERS-210s last year. Our gut feeling is that | newer versions of the OPEN-R SDK are the cause. That said, streaming data from | the robot using TCP (and communicating between robots using UDP) seems to work | flawlessly ! | | We have experimented with a number of different versions of TinyFTPD and they | all seem to have the problem. Is anyone else experiencing this or are we doing | something weird ? | |
| |- 1060 RE:TinyFTPD peterleko [2004/04/23-02:40] |
| | Response : Edit : Delete |
|
| Not in my team. The ftp works fine. | | >We've been experiencing problems using TinyFTPD on our ERS-7s (haven't tested | >ERS-210s). Basically, it's sometimes very very slow to upload (and sometimes | >doesn't work at all), even for small files (~1kb or less). Other times it | fails | >to upload the entire file, particularly for larger files (~100kb). | > | >We didn't have this problem with our ERS-210s last year. Our gut feeling is | that | >newer versions of the OPEN-R SDK are the cause. That said, streaming data from | >the robot using TCP (and communicating between robots using UDP) seems to work | >flawlessly ! | > | >We have experimented with a number of different versions of TinyFTPD and they | >all seem to have the problem. Is anyone else experiencing this or are we doing | >something weird ? | |
| +- 1109 RE:TinyFTPD GriffithJF [2004/05/26-15:07] |
| Response : Email : Edit : Delete |
|
We
are having similar problems on some of our machines. one thing that may eb significant if you are using a mixed environment: The object determines eol chars based on the machine it was compiled on. Specifically if you compile it on a windows machine and connect from unix (or vice versa) you may have problems. |
| V 1068 Reading distance sensors with ERS-7 using OPEN-R outsider [2004/04/27-23:46] |
| | Response : Edit : Delete |
|
| Hi, | I would like to read the value of the distance sensors on a ERS-7 model. The | problem is that the 'Model Information' document says that 'values for the | near/far distance sensors appears alternatively'. So I don't understand how can | I decide if the value read belongs to the near sensor or to the far sensor | (since they don't show the same value on reading). | My problem arises because I want to have a PC program that shows on the screen | the following: | | Near distance sensor : 5 cm | Far distance sensor : 12 cm | | For that reason I need to know to which sensor the measurement belongs to. | | Thank you | |
| +- 1071 RE:Reading distance sensors with ERS-7 using OPEN-R TodboT [2004/04/29-01:40] |
| | Response : Edit : Delete |
|
| The
document may be a little misleading. Take a look at the SensorObserver7 | sample. The near & far PSD both have their own CPC locator, so they can be | accessed individually. However, the near/far PSD is alternately updated. You | can also look at the 'signal' value. Only one of the PSDs will have a valid | signal at one time. The 'value' remains latched if the other PSD is valid. | |
| +- 1081 RE:Reading distance sensors with ERS-7 using OPEN-R outsider [2004/05/12-16:30] |
| | Response : Edit : Delete |
|
| Thank
You TodBot for your answer. It now makes sense to me. I haven't had time | yet to test what you say but I'll do it in short and make a post here for | anybody else who could have the same missconception. | By the way, are you the same TodBot as in AIBO-Life Forum?. I am the same | outsider... | |
| +- 1084 RE:Reading distance sensors with ERS-7 using OPEN-R TodboT [2004/05/14-00:49] |
| | Response : Edit : Delete |
|
| One
and the same. | | >By the way, are you the same TodBot as in AIBO-Life Forum?. | |
| +- 1102 RE:Reading distance sensors with ERS-7 using OPEN-R outsider [2004/05/24-16:10] |
| Response : Edit : Delete |
|
Thanks
for your explanation. Now it works!. Arigato gozaimashita!! |
| V 1096 [Q] Questions on Programming facilities of ERS-7 xhiloh [2004/05/22-00:50] |
| | Response : Email : Edit : Delete |
|
| Hello all, | | I'm considering whether ERS is worth purchase | So I'm looking into the programming facilities of AIBO | | Though I read FAQs, | I still have some questions | | 1) Can I use openr sdk and motion editor in cooperative manner? | That is, can I make motion data for openr system calls using | motion editor? Or should I create all motion data manually? | | 2) Assume that I use openr sdk. | What facilities can I use when I make voice detection | or object detection application ? | | For example, color detection table or special predefined | function for image processing, etc | | 3) Can I use openr and rcode in cooperative manner? | That is, is it possible that one part of a program is constructed | by openr and the other part of the program is constructed by rcode? | | 4) When I use rcode, should I make scripts for each objects? | If so, can I set priority for each object? | | 5) In openr, can one manipulate the status of objects dynamically? | That is, is it possible that one can stop or suspend an object | at a moment, and restart it some times later? | | I long for your answers | Please answer to the questions | | Thank you. | (The article that 2004/05/22-00:48 posted, was modified by Author) | |
| +- 1098 RE:[Q] Questions on Programming facilities of ERS-7 AiboPet [2004/05/23-00:21] |
| | Response : Edit : Delete |
|
| >1)
Can I use openr sdk and motion editor in cooperative manner? | > That is, can I make motion data for openr system calls using | > motion editor? Or should I create all motion data manually? | Yes, you can do both. | The new MEdit motion editor (or the free Skitter motion editor or other | editors for older models) all use the same .MTN motion format (.MTN files are | usually combined together into one big .ODA file) | .MTN playback is built-in to RCODE. | .MTN playback is not built-in to the OpenR SDK, but it is provided as part of | the samples (include the MoNet module and glue it together). See the | BallTrackingHead7 sample for one way to do this. | | >2) Assume that I use openr sdk. | > What facilities can I use when I make voice detection | > or object detection application ? | There is no facility for voice detection in the OpenR SDK. You are on your own. | Implementing your own sound detection or tone detection should be rather easy - | - accessing the raw audio data (voice is a lot harder of course) | | RCODE has voice detection built-in with a pre-defined set of phrases (53 of | them). That's one of the main reasons why RCODE is better for "entertainment" | personalities IMHO. | | There is no formal object detection in the OpenR SDK. There are several | samples that illustrate different techniques for color object detection | (BallTrackingHead7 is the best one) | The RoboCup teams have much more elaborate object detection for playing soccer | on the field. | | > For example, color detection table or special predefined | > function for image processing, etc | | Color detection is handled very well (built into the hardware, see samples and | API). | For the rest of the stuff you are mostly on your own (or need to glue together | parts from RoboCup source code) | | >3) Can I use openr and rcode in cooperative manner? | > That is, is it possible that one part of a program is constructed | > by openr and the other part of the program is constructed by rcode? | | No. The two worlds are very separate. | If you think removing this limitation would be a good idea, please post a | special request on this BBS [eventually Sony may change their minds, but don't | hold your breath ;-] | | One compromise is to have part in RCODE, and part in C++ code running on a | WiFi connected PC (no OpenR SDK). | | >4) When I use rcode, should I make scripts for each objects? | > If so, can I set priority for each object? | | RCODE has one single script. It is not object oriented. | [ie. one big procedural program, with subroutines, global variables etc] | | You can structure it anyway you want for encapsulation purposes (like old | style BASIC code). There is also a set of RCODE tools by John of DogsBody that | includes an RCODE script pre-processor. | http://www.dogsbodynet.com/aibnet.html | | >5) In openr, can one manipulate the status of objects dynamically? | > That is, is it possible that one can stop or suspend an object | > at a moment, and restart it some times later? | | It depends what you want to do. | | IMHO: The generally documented features of the object model used in OpenR are | a means to an end. You encapsulate your code in objects instead of in separate | DLLs. | I like to think of them more like separate mini-processes, not C++ objects. | | For example, you don't have multiple instances of the same object, no real | inheritance, so the "object orientation" is somewhat limited (just for process | encapsulation and messaging) | Most of the wiring is hard coded, but there are a few undocumented OpenR/COOP | APIs for playing around with the lower level plumbing. | | Also look at the Tekkotsu project. They have a more elaborate framework | layered on top of OpenR (ie. they deal with the limitations of the OpenR | object model for you, and provide a more C++-like object model with templates | and inheritance) | http://www-2.cs.cmu.edu/~tekkotsu/ | | Hope that yelps | |
| +- 1100 RE:[Q] Questions on Programming facilities of ERS-7 xhiloh [2004/05/24-11:35] |
| Response : Email : Edit : Delete |
|
Your
answer helps much. Thank you :) |
| V 1099 [Q] Can't run Motion Editor jkp3277 [2004/05/23-21:02] |
| Response : Email : Edit : Delete |
|
I
successfully downloaded and installed AIBO Motion Editor. However, MEditexe is not run well. Whenever I click the MEdit.exe, error reporting popup window showing error message like "MEdit MFC Application have a problem" is just raised. My computer is as follows: - Korean Windows XP with SP1 - Sufficient memory and HD - Matrox Millennium G AGP with MB Urgently help me, please. (The article that 2004/05/23-10:39 posted, was modified by Author) |
| V 1061 AIBO newbie and Java programmer javajuster [2004/04/25-06:08] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hello all. | | I am a brand new AIBO owner as of Monday this week. I'm having a lot of fun | figuring out what he can do! I am also a Java programmer, and was intrigued to | learn that I could program AIBO to do things with the OPEN-R SDK. However, I | am a C++ novice to say the least. Also, I am used to the Windows platform | mostly but do have some LINUX/UNIX experience. | | Could someone please suggest to me how to begin developing for AIBO? Do you | recommend I spin up a LINUX machine to save me the possible grief with Cygwin | (something I've heard horror stories about)...? Also, what experience has | anyone had, if any, programming C++ from a primarily Java background. I feel | like I've been spoiled with some of the niceties that Java provides when it | comes to memory management, references, etc. | | Any pointers from the group? | |
| |- 1093 RE:AIBO newbie and Java programmer harrison [2004/05/21-03:20] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Hello all. | > | >I am a brand new AIBO owner as of Monday this week. I'm having a lot of fun | >figuring out what he can do! I am also a Java programmer, and was intrigued to | >learn that I could program AIBO to do things with the OPEN-R SDK. However, I | >am a C++ novice to say the least. Also, I am used to the Windows platform | >mostly but do have some LINUX/UNIX experience. | > | >Could someone please suggest to me how to begin developing for AIBO? Do you | >recommend I spin up a LINUX machine to save me the possible grief with Cygwin | >(something I've heard horror stories about)...? Also, what experience has | >anyone had, if any, programming C++ from a primarily Java background. I feel | >like I've been spoiled with some of the niceties that Java provides when it | >comes to memory management, references, etc. | > | >Any pointers from the group? | | Linux in general is useful to know how to use, as for you and your aibo, it is | an asset. There is many tutorials in this site that you can use to get used to | the aibo environment. The most important thing you should do right now is | getting your aibo setup for development. Make sure you have a good memory | stick reader (the wrong one can corrupt the memory stick permanently; its | happened to me before, very frustrating) with a programmable memory stick, and | a cygwin setup. Once you get the environment setup on your computer, download | the samples, and start compiling those and testing them out. Get an idea about | how aibo works and the directory structure. | | Those are all the basics, after that you will feel more comfortable with | designing and implementing a successful aibo program. Good luck. | |
| +- 1095 RE:AIBO newbie and Java programmer AiboPet [2004/05/22-00:11] |
| Response : Edit : Delete |
|
>I
am a brand new AIBO owner as of Monday this week.... First suggestion: Stick with Windows. The learning curve for C++ and the OPEN- R SDK is already pretty steep. CYGWIN works great for me (under Windows XP). Follow the installation instructions exactly, and be sure you don't have older/other CYGWIN installations on the machine. Once you are in the CYGWIN shell, it is very similar to a simplified Unix/Linux shell. You can also run Windows programs, .BAT files etc. Once you get into the habit of typing "make" on the command line, the actual operating system doesn't matter as much. Second suggestion: If you haven't already, look at RCODE (a script-like programming language) Now available for the ERS-7, also available for all the other model AIBOs It depends on the kind of robot programming you want to create: IMHO: For robot experiments (like custom walking), the OpenR SDK is great For creating entertaining "personalities", using RCODE is much easier. Third suggestion/plug: For even easier/faster results, look at YART: http://www.aibohack.com/rcode/yart.htm http://www.aibohack.com/ers7/yart7.htm (for ERS-7, new) YART is a tool I wrote that makes it very easy to get into RCODE programming. I always recommend people start with YART, even if you have a long term goal of doing hard core programming with the SDK [YART is the gateway drug to RCODE, RCODE is the gateway drug to the OpenR SDK ;=] |
| V 0819 Need help... No boot.. harrison [2003/11/20-12:49] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hey everyone, | | I recenlty got an ERS-210A and wanted to program some stuff using the open-r | sdk. Currently, I have open-r compiling all the sample programs without errors | under windows using cygwin. One I throw the compiled sample along with the | base "BASIC" open-r tree. Once i get the stick into the dog, it doesn't boot | correctly, it paused for about 10 seconds before it plays a negative melody and | shuts off. | My guess is that I either formated the memory stick wrongly (although I could | have swore I did a quick format) or there is something else I am missing. | However, if the memory stick was formatted, is it possible to do some type of | imageing (using dd in linux) process that can be done to make the stick useable | again? | | -Thanks | |
| +- 0820 RE:Need help... No boot.. AiboPet [2003/11/21-00:49] |
| | Response : Edit : Delete |
|
| >...
I have open-r compiling all the sample programs without errors | >under windows using cygwin. One I throw the compiled sample along with the | >base "BASIC" open-r tree. Once i get the stick into the dog, it doesn't boot | >correctly, it paused for about 10 seconds before it plays a negative melody | and shuts off.... | | Common advice: | What model AIBO ? The latest SDK (1.1.5) only works with the ERS-7, the | slightly older SDK (1.1.4) only works for the ERS-2x0. | | If you have a very old ERS-210, you may need to install the Flash ROM upgrade | (available under downloads) before you can run any OpenR SDK programs | | You need to use a "PMS" programmable memory stick for OpenR SDK programs. If | the stick has an "ERF-" model number sticker on it, it is not a PMS stick. | | If you have an ERS-2x0 (and you have an updated Flash ROM), try a known | working program on the PMS stick, like Poser: | http://www.aibohack.com/openr_sdk/index.html#POSER | Follow the instructions there. If this fails to boot, then most likely the | copy protection is gone on the pink "PMS" memory stick. | | >... if the memory stick was formatted, is it possible to do some type of | >imageing (using dd in linux) process that can be done to make the stick | useable again? | | No. | What brand memory stick reader are you using ? | Have you used it successfully before with Pink sticks ? | | [in general I only recommend Sony brand stick readers - other brands may wipe | out memory sticks in a second if you put them in the stick reader - even with | the write protect switch on] | |
| +- 0824 RE:Need help... No boot.. harrison [2003/11/21-06:10] |
| | Response : HomePage : Email : Edit : Delete |
|
| >>...
I have open-r compiling all the sample programs without errors | >>under windows using cygwin. One I throw the compiled sample along with the | >>base "BASIC" open-r tree. Once i get the stick into the dog, it doesn't boot | >>correctly, it paused for about 10 seconds before it plays a negative melody | >and shuts off.... | > | >Common advice: | >What model AIBO ? The latest SDK (1.1.5) only works with the ERS-7, the | >slightly older SDK (1.1.4) only works for the ERS-2x0. | | I got OPEN_R_SDK-sample-1.1.4-r1 working now | > | >If you have a very old ERS-210, you may need to install the Flash ROM upgrade | >(available under downloads) before you can run any OpenR SDK programs | | I have updated the flash although I have an ERS-210A. When i was reading up on | that, it said it wouldn't even work if the flash didn't need to be updated, so | i tried it anyways. | > | >You need to use a "PMS" programmable memory stick for OpenR SDK programs. If | >the stick has an "ERF-" model number sticker on it, it is not a PMS stick. | | Yup, i got the Pink Sony open-r programmable memory stick. | > | >If you have an ERS-2x0 (and you have an updated Flash ROM), try a known | >working program on the PMS stick, like Poser: | >http://www.aibohack.com/openr_sdk/index.html#POSER | >Follow the instructions there. If this fails to boot, then most likely the | >copy protection is gone on the pink "PMS" memory stick. | | I've actually tried this already with no dice when i was searching the forums. | > | >>... if the memory stick was formatted, is it possible to do some type of | >>imageing (using dd in linux) process that can be done to make the stick | >useable again? | > | >No. | >What brand memory stick reader are you using ? | >Have you used it successfully before with Pink sticks ? | > | >[in general I only recommend Sony brand stick readers - other brands may wipe | >out memory sticks in a second if you put them in the stick reader - even with | >the write protect switch on] | | Well, thanks for your help. However, I believe that it could be that the flash | is corrupted. I tried booting the dog with no memory stick inside, and it | fails to boot. It only boots now with the help of the Aibo eyes memory stick, | the one thing i haven't touched. | | Any ideas? | | Thanks again, | -Harrison | |
| +- 0827 RE:Need help... No boot.. AiboPet [2003/11/22-01:51] |
| | Response : Edit : Delete |
|
| >
I got OPEN_R_SDK-sample-1.1.4-r1 working now... | | I assume you mean compiling, but not actually running on AIBO. | | > I have updated the flash although I have an ERS-210A.... | | As you point out, it should not be needed. | Did it do anything ? This may be the cause of the theorized flash corruption | | >> If you have an ERS-2x0 (and you have an updated Flash ROM), try a known | >>...working program on the PMS stick, like Poser: | > I've actually tried this already with no dice when i was searching the | forums. | | That is bad. Either the stick is bad, or the AIBO is bad (or both) | | > However, I believe that it could be that the flash | > is corrupted. I tried booting the dog with no memory stick inside, and it | > fails to boot. | | That is very bad. Did this happen after attempting the flash upgrade ? | If that far gone, the only solution may be sending him back to Sony for repair. | | > It only boots now with the help of the Aibo eyes memory stick, | > the one thing i haven't touched. | | What brand memory stick reader are you using ? | | I suspect you have two problems: a memory stick reader that corrupts the copy | protection, and a potential flash memory problem. | | NOTE: AiboEyes and some of the other newer sticks have most/all of the SYSTEM | modules loaded from memory stick, so they rely less on the built-in flash | memory (compared to AiboLife or RCODE programs for example). However, the | OpenR SDK does this as well, so that's why I suspect two separate problems. | |
| +- 0829 RE:Need help... No boot.. harrison [2003/11/26-06:39] |
| | Response : HomePage : Email : Edit : Delete |
|
| >>
I got OPEN_R_SDK-sample-1.1.4-r1 working now... | > | >I assume you mean compiling, but not actually running on AIBO. | | yup, | | >> I have updated the flash although I have an ERS-210A.... | > | >As you point out, it should not be needed. | >Did it do anything ? This may be the cause of the theorized flash corruption | | The update went well, it gave the beeps that it normal would do in the event | that it needed to be updated. | > | >>> If you have an ERS-2x0 (and you have an updated Flash ROM), try a known | >>>...working program on the PMS stick, like Poser: | >> I've actually tried this already with no dice when i was searching the | >forums. | > | >That is bad. Either the stick is bad, or the AIBO is bad (or both) | | I have tried popping in the aibo eyes memory stick into my Dazzel PCMCIA memory | stick reader, but I haven't been able to see anything on it. It asks me if i | want to format, and i know i dont want to do that. However, is it possible for | the reader to automatically format those magic bytes of copy protection without | my so-called consent? | > | >> However, I believe that it could be that the flash | >> is corrupted. I tried booting the dog with no memory stick inside, and it | >> fails to boot. | > | >That is very bad. Did this happen after attempting the flash upgrade ? | | Yup | | >If that far gone, the only solution may be sending him back to Sony for repair. | > | >> It only boots now with the help of the Aibo eyes memory stick, | >> the one thing i haven't touched. | > | >What brand memory stick reader are you using ? | Dazzle 4-in-1 pcmcia card adapter | > | >I suspect you have two problems: a memory stick reader that corrupts the copy | >protection, and a potential flash memory problem. | | But then again, I have copied and ran the update from the memory stick from | this reader which makes me feel that the reader isn't neccesarily the problem. | However, I am willing to go out and buy a sony brand memory stick reader too, | thats a small extra investment in the large scheme of things. | | About the repairs at sony, how would i go about sending my Aibo there so that I | can be able to feel that it would be taken care of safely and quickly? It | makes me alittle nervous. | | Thanks for your returning help it is most appreciated esp. in my great time of | need, | -Harrison | > | >NOTE: AiboEyes and some of the other newer sticks have most/all of the SYSTEM | >modules loaded from memory stick, so they rely less on the built-in flash | >memory (compared to AiboLife or RCODE programs for example). However, the | >OpenR SDK does this as well, so that's why I suspect two separate problems. | |
| +- 0830 RE:Need help... No boot.. AiboPet [2003/11/27-02:33] |
| | Response : Edit : Delete |
|
| Selective
replies re stick readers | | >I have tried popping in the aibo eyes memory stick into my Dazzel PCMCIA | memory | >stick reader, but I haven't been able to see anything on it. It asks me if i | >want to format, and i know i dont want to do that. | | That's a sign to never use that stick reader again ! | Hopefully the stick is still ok. | | > However, is it possible for | >the reader to automatically format those magic bytes of copy protection | without | >my so-called consent? | | Yes. | In one especially evil case of a non-Sony reader, placing a stick into the | memory stick reader for one second immediately corrupted the stick, without | doing anything. Even with the PROTECT switch turned on! | | Not all stick readers are that evil, but there are many cases of non-Sony | brand readers wiping the copy protection. I have never seen/heard of a case | where a Sony brand reader caused this problem (even the stick readers in Sony | cameras or PDAs are ok) | | >However, I am willing to go out and buy a sony brand memory stick reader too, | >thats a small extra investment in the large scheme of things. | | Agreed | Some people have had luck with non-Sony brand readers, but for the cost of a | Sony brand stick reader, it is better to be safe. | | I personally like the brand new MSAC-US30 (USB/USB2, a littly pricey). It is | MS Pro compatible but also faster than older readers for normal memory sticks. | | ---- | >About the repairs at sony, how would i go about sending my Aibo there so that | I | >can be able to feel that it would be taken care of safely and quickly? It | >makes me alittle nervous. | | There are anecdotal stories that run the gauntlet of really good experiences | to really bad experiences. | | Many of the bad experiences are when the user thinks there is a problem, and | Sony service technicians don't. The fact it can't boot with the memory stick | slot empty is clearly a sign there is something wrong. The replacable part | (the main body brick) is relatively easy to replace | | I'd try a few things with a Sony brand stick reader first (ie. try to get any | of your PMS sticks to work). | |
| +- 1094 RE:Need help... No boot.. harrison [2004/05/21-03:25] |
| Response : HomePage : Email : Edit : Delete |
|
I
sent the stick back to sony and got it reformatted along with buying
the sony brand name reader. Everything is cool now =) Thanks for your help (sorry for delayed response, been busy with schoolin') |
| V 1088 How to delay for few time units without having full CPU use? wangy22 [2004/05/19-01:03] |
| | Response : Edit : Delete |
|
| I am
new to OPEN-R SDK and trying to implement video streaming from AIBO to
PC. | I want AIBO to grab the image every 40ms. How do can I delay it 40ms without | having AIBO's CPU fully used? I tried to have a loop keep checking the | timeDifference from OPENR::GetTimeDifference(), but I think it's a busy waiting. | | Thanks! | |
| +- 1089 RE:How to delay for few time units without having full CPU use? ejt [2004/05/19-02:11] |
| | Response : HomePage : Email : Edit : Delete |
|
| >I
am new to OPEN-R SDK and trying to implement video streaming from AIBO
to PC. | >I want AIBO to grab the image every 40ms. How do can I delay it 40ms without | I'm afraid you've got it a little backwards - you don't want to delay for 40ms, you want to return to the | system from the last frame within 40ms. | The system pushes frames to your stub function with OPEN-R messages, and it won't send you a new | frame while you still haven't returned from the processing on the previous frame. I wonder tho, if you | take long enough, if the system will run out of buffers and rewrite the image buffer underneath you, | even tho you are still "processing" it. I'd be surprised however. | |
| +- 1090 RE:How to delay for few time units without having full CPU u wangy22 [2004/05/19-02:41] |
| | Response : Edit : Delete |
|
| Are
you saying that the duration of transferring NotifyEvent.data from
Camera | subject to the data observer might be longer than 40ms? If yes, how can I solve | this problem? | | I really have no experience in implementing video streaming, do you have any | suggestions? | | Thanks a lot! | | >>I am new to OPEN-R SDK and trying to implement video streaming from AIBO to PC. | >>I want AIBO to grab the image every 40ms. How do can I delay it 40ms without | >I'm afraid you've got it a little backwards - you don't want to delay for 40ms, | you want to return to the | >system from the last frame within 40ms. | >The system pushes frames to your stub function with OPEN-R messages, and it | won't send you a new | >frame while you still haven't returned from the processing on the previous | frame. I wonder tho, if you | >take long enough, if the system will run out of buffers and rewrite the image | buffer underneath you, | >even tho you are still "processing" it. I'd be surprised however. | |
| +- 1091 RE:How to delay for few time units without having full CPU u ejt [2004/05/19-10:52] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Are
you saying that the duration of transferring NotifyEvent.data from
Camera | >subject to the data observer might be longer than 40ms? If yes, how can I solve | >this problem? | No, although I don't know what the lag time is between the subject sending and an observer receiving | is. That actually used to be rather significant, but I believe they have greatly reduced this in more | recent versions of OPEN-R. | | What I am saying is that if you want to receive frames at 'n' frames per second, then you must finish | processing of each frame in less than 1/n seconds. Unless you change your queue settings | (SetBufCtrlParam()), a subject will not send an observer a new message if that observer is still doing | processing on the previous message. Instead, it just drops the message and throws away the data. | | SetBufCtrlParam() can set it to queue up frames instead of dropping them, but if you're not finishing | each frame in less than 1/n seconds, you're still going to wind up dropping frames after you blow the | queue. | | So, if you set up a loop in the observer, hoping to busy wait until the next message, you'll loop forever. | You need to return from the observer all together, which lets the system know you are done processing | and ready for more information. | |
| +- 1092 RE:How to delay for few time units without having full CPU u wangy22 [2004/05/19-11:11] |
| Response : Edit : Delete |
|
OK,
I don't mind dropping frames. So what is the time unit for skip in
SetSkip()? Thanks! >>Are you saying that the duration of transferring NotifyEvent.data from Camera >>subject to the data observer might be longer than 40ms? If yes, how can I solve >>this problem? >No, although I don't know what the lag time is between the subject sending and an observer receiving >is. That actually used to be rather significant, but I believe they have greatly reduced this in more >recent versions of OPEN-R. > >What I am saying is that if you want to receive frames at 'n' frames per second, then you must finish >processing of each frame in less than 1/n seconds. Unless you change your queue settings >(SetBufCtrlParam()), a subject will not send an observer a new message if that observer is still doing >processing on the previous message. Instead, it just drops the message and throws away the data. > >SetBufCtrlParam() can set it to queue up frames instead of dropping them, but if you're not finishing >each frame in less than 1/n seconds, you're still going to wind up dropping frames after you blow the >queue. > >So, if you set up a loop in the observer, hoping to busy wait until the next message, you'll loop forever. >You need to return from the observer all together, which lets the system know you are done processing >and ready for more information. |
| V 1087 The R-CODE SDK and AIBO Motion Editor have been released. openrsupport@(Administrator) [2004/05/18-01:19] |
| Response : Edit : Delete |
|
The
R-CODE SDK and AIBO Motion Editor have been released. You can find them on the download page. https://openr.aibo.com/openr/eng/perm/download.php4 AIBO SDE Support |
| V 1079 How are RCRegion, OShmPtrBase and OShmPtr related? wangy22 [2004/05/12-05:10] |
| | Response : Edit : Delete |
|
| Could
anyone tell me what are the relations between RCRegion, OShmPtrBase and | OShmPtr? What is MemID? Are there any documentations other than the reference | guide talking about these concepts? | | Thanks for help! | |
| +- 1086 RE:How are RCRegion, OShmPtrBase and OShmPtr related? AiboPet [2004/05/16-01:36] |
| Response : Edit : Delete |
|
>Could
anyone tell me what are the relations between RCRegion, OShmPtrBase and >OShmPtr? What is MemID? Are there any documentations other than the reference >guide talking about these concepts? These are mostly undocumented internals of OpenR. You can figure most of them out by looking at the headers. MemID (MemoryRegionID) is the lowest level memory allocation chunk (like a memory handle in other operating systems) - for sharing memory between processes. RCRegion is a C++ class that encapsulates a MemID and includes other features like reference counting and memory offset addressing OShmPtr is a C++ template class that encapsulates a MemID and includes C++ type-safe access to the data OShmPtrBase is used to help implement OShmPtr My general advice is to use them when needed for other APIs (usually RCRegion when calling SetData etc), and ignore them otherwise. Inside your own program, use any memory allocation technique you want (eg: normal C++ operator new/delete) |
| V 1050 How to maintain Aibo powered up all the time outsider [2004/04/13-17:04] |
| | Response : Email : Edit : Delete |
|
| Hi, | I'm a Ph.D. researcher doing some evolutionary robotics experiments. I would | like to use my Aibo ERS-7 robot for those experiments. Unfortunately, the | experiments take place over a large period of time, so battery life is an issue. | I would like to know if it would be possible to have Aibo powered during large | periods of time in some way (even if this requires special electronic hardware). | Can I buy any special connector that plugs Aibo directly to AC while moving | arround?. Can I construct it by myself?. | | Domo arigato. | |
| +- 1054 RE:How to maintain Aibo powered up all the time openrsupport@(Administrator) [2004/04/21-00:37] |
| | Response : Edit : Delete |
|
| I'm
sorry, but there is no special connector for an AIBO ERS-7 that Sony
sells. | It is not allowed to have AIBO operate for very long periods of time with lots | of movement. AIBO would generate a lot of heat, this would increase the wear on | parts, and AIBO can be damaged with prolonged exposure to high temperatures. | | OPEN-R SDK Support | |
| +- 1066 RE:How to maintain Aibo powered up all the time outsider [2004/04/27-23:38] |
| | Response : Edit : Delete |
|
| Thank
you very much for your answer. | By the way, which temperature will you consider high temperature for Aibo?. For | how long could Aibo run at that temperature?. | (I refer to the internal temperature of the core that can be read using Open-r | command OPENR::GetPowerStatus ). | | Thank you | |
| +- 1074 RE:How to maintain Aibo powered up all the time openrsupport@(Administrator) [2004/05/07-23:56] |
| | Response : Edit : Delete |
|
| AIBO
(OPEN-R SDK System objects) automatically detect high temperatures and | shutdown the robot. | | The thresholds are: | | Shutdown threshold for battery discharging: | ERS-210/220 70 degrees Celsius battery temperature | ERS-7 65 degrees Celsius battery temperature | | Shutdown threshold for battery charging: | ERS-210/220 65 degrees Celsius battery temperature | ERS-7 60 degrees Celsius battery temperature | | AIBO isn't designed to operate with the A/C adapter while a load is placed on | each joint (walking). AIBO's cooling system (hardware design) can't cope with | this situation. If you want to continually supply power to AIBO with the A/C | adapter, please use AIBO on the Energy Station. | | Thank you, | | OPEN-R SDK Support | |
| +- 1080 RE:How to maintain Aibo powered up all the time outsider [2004/05/12-16:23] |
| Response : Edit : Delete |
|
Thank
you! |
| V 1078 Presentation Data from OPEN-R SIG and FAQ C0010 openrsupport@(Administrator) [2004/05/12-00:27] |
| Response : Edit : Delete |
|
Hi everyone, The presentation data from the OPEN-R SIG at the RoboCup American Open 2004 can now be downloaded in the OPEN-R SDK University section of this website. FAQ C0010 has been added. Q. Can I prevent AIBO's shutdown due to 'BATTERY_OVER_CURRENT'? Thank you, OPEN-R SDK Support |
| V 1075 Why must I rebuild GCC 3.3? Mac OS X 10.3.x already has it. Conholio [2004/05/10-00:27] |
| | Response : Edit : Delete |
|
| I
have xCode already installed on my Mac which uses GCC 3.3. Why is it
required to download and | rebuild it to setup the Open-R SDK? Is there Mac specific instructions somewhere for my situation? | | | |
| +- 1076 RE:Why must I rebuild GCC 3.3? Mac OS X 10.3.x already has it. ejt [2004/05/10-17:23] |
| | Response : HomePage : Email : Edit : Delete |
|
| The
installed gcc is built to target a PowerPC processor, using libc and
binutil also targeting PowerPC. | | But, as you may guess, the AIBO uses a different chip (MIPS) so you have to build a copy of gcc | configured to cross-compile to target the MIPS chip instead. (and you'll need newlib and a different | binutil to go with it) | | Just follow the instructions for Linux, except grab the Mac OS X build script instead of the linux one. | it's fairly straightforward as these things go. (i've got it installed on my iBook) I think you might have to | settle for using make and the commandline to build tho. If there's a way to configure xCode to use the | OPEN-R installation, that would be interesting. Not sure if it'd be worthwhile tho, can just use the | editor and then just command-tab and type 'make' (or up-arrow and return) | | -ethan | |
| +- 1077 RE:Why must I rebuild GCC 3.3? Mac OS X 10.3.x already has it. Conholio [2004/05/11-00:18] |
| Response : Edit : Delete |
|
I
gathered that it was for cross compiling after I posted. Now my system
can x-compile for aibo and my zaurus, as well as use x-code's debugger for qtopia apps. I think you can rig x-code to do cross-compilations and use custom file types/compilers but I haven't delved to deep. Thanks for the response. >The installed gcc is built to target a PowerPC processor, using libc and binutil also targeting PowerPC. > >But, as you may guess, the AIBO uses a different chip (MIPS) so you have to build a copy of gcc >configured to cross-compile to target the MIPS chip instead. (and you'll need newlib and a different >binutil to go with it) > >Just follow the instructions for Linux, except grab the Mac OS X build script instead of the linux one. >it's fairly straightforward as these things go. (i've got it installed on my iBook) I think you might have to >settle for using make and the commandline to build tho. If there's a way to configure xCode to use the >OPEN-R installation, that would be interesting. Not sure if it'd be worthwhile tho, can just use the >editor and then just command-tab and type 'make' (or up-arrow and return) > >-ethan |
| V 1072 monet commands mrbrand [2004/05/03-23:52] |
| | Response : Email : Edit : Delete |
|
| where
can i find a list of commands to use in the Monet sample? | | i've searched the internet and all i found was "shutdown". (which worked btw.) | some numbers seem to work too: | "2" makes AIBO stand up, | "1" sit dowm, | "0" lay down etc.. | | but there have got to be more... | |
| +- 1073 RE:monet commands AiboPet [2004/05/04-01:22] |
| Response : Edit : Delete |
|
>where
can i find a list of commands to use in the Monet sample? > >i've searched the internet and all i found was "shutdown". (which worked btw.) "shutdown" is a special case (in the code) The other command numbers are specified in the "MONETCMD.CFG" file (on the memory stick (in \OPEN-R\MW\CONF or a subfolder) >...but there have got to be more... There aren't too many more. This is a simple sample. If you want more, you have to add them yourself |
| V 1069 Problem executing ObjectComm(Host) from Release 2 angel [2004/04/28-21:02] |
| | Response : HomePage : Email : Edit : Delete |
|
| After
installing the release 2 of Open-R, I went to test some of the sample | apps. A number of them work, but I have a problem with ObjectComm when I run it | in host mode on my cygwin console. | | (1) Goto common/ObjectComm/host | (2) make install | (3) run ipc-daemon2 in the background | (4) goto /usr/local/OPEN_R_SDK/RP_OPEN_R/bin | (5) execute start-rp-openr | | Here is the output | | miguel@venus /usr/local/OPEN_R_SDK/RP_OPEN_R/bin | $ ./start-rp-openr | [pid:3076,msqid:2,oid:0x00000002] oserviceManager | [pid:3156,msqid:3,oid:0x00000003] tcpGateway | MS/OPEN-R/MW/CONF/OBJECT.CFG: No such file or directory | | I checked that the OBJECT.CFG exists in the MS folder under ObjectComm/host | Any ideas? | |
| +- 1070 RE:Problem executing ObjectComm(Host) from Release 2 TodboT [2004/04/29-00:34] |
| Response : Edit : Delete |
|
I
haven't used RP that much, but I think you need to stay in
ObjectComm/RP/host and then execute start-rp-openr. Something like this: (1) Goto common/ObjectComm/RP/host (2) make install (3) run ipc-daemon2 in the background (4) execute /usr/local/OPEN_R_SDK/RP_OPEN_R/bin/start-rp-openr |
| V 1062 Welcome to the AIBO SDE openrsupport@(Administrator) [2004/04/26-12:33] |
| | Response : Edit : Delete |
|
| Welcome to our refreshed website! | | We have greatly expanded the available tools for | AIBO programming, and have called the expanded | suite "AIBO SDE", which stands for AIBO Software | Development Environment. The AIBO SDE consists of | the OPEN-R SDK, the R-CODE SDK, the AIBO Remote | Framework, and the AIBO Motion Editor. | | Of course, the OPEN-R SDK is available now. The | other new tools will be released soon. In the meantime, | please read the extensive FAQs to learn more about | the new tools. | | We hope that you enjoy the new site and that the | new tools will greatly benefit your work. | | Thank you, | | AIBO SDE Support | |
| |- 1065 RE:Welcome to the AIBO SDE stainlessjimmy [2004/04/27-17:10] |
| | Response : Edit : Delete |
|
| Great ! | Many thanks to SONY - Developer ! Now there are much more reasons to buy an | AIBO then ever ! | | Jimmy | |
| +- 1067 RE:Welcome to the AIBO SDE outsider [2004/04/27-23:40] |
| Response : Edit : Delete |
|
Congratulations
for your hard work!. I know how difficult is to develop such kind of tools. We all thank you for releasing those. Regards |
| V 1063 memory stick programmer mrbrand [2004/04/26-19:22] |
| | Response : Email : Edit : Delete |
|
| i am
a total noob at programming the aibo and i have a totat nooby question: | | until recently i thought i could transfer the open-r code directly to the stick | through the wireless lan without ever using a memorystick programmer. | | but that not possible, is it? | |
| +- 1064 RE:memory stick programmer AiboPet [2004/04/27-01:25] |
| Response : Edit : Delete |
|
>until
recently i thought i could transfer the open-r code directly to the stick >through the wireless lan without ever using a memorystick programmer. This is a common myth. Once most of the base software is installed and net configured, you can do incremental changes with TinyFTPD, like uploading new .BIN modules and OBJECTS.CFG and other data files. The updated files get stored on the memory stick. Then after you reboot your AIBO it will run the software now on the memory stick. However, for everything else including the initial setup, LAN configuration, etc, you will need a stick reader. So you need both: a memory stick reader (SONY brand strongly recommended) and a WiFi connection to your PC. |
| V 1048 OPEN-R SDK 1.1.5-r2 Has Been Released openrsupport@(Administrator) [2004/04/13-01:44] |
| | Response : Edit : Delete |
|
| Hi everyone, | | OPEN-R SDK 1.1.5-r2 has been released. The high-current detection threshold in | the OPEN-R SDK 1.1.5 r2 has been changed and ERS-7 is less likely to shut down. | Battery plunger is locked during startup and shutdown for EEPROM protection. | | Thank you, | | OPEN-R SDK Support | |
| +- 1057 RE:OPEN-R SDK 1.1.5-r2 Has Been Released gijoe [2004/04/22-16:05] |
| | Response : HomePage : Edit : Delete |
|
| Thanks
for this ! Appears to work fine. We haven't seen a BATTERY_OVER_CURRENT | since, including at the Australian Open. | | -NUbots. | |
| +- 1059 RE:OPEN-R SDK 1.1.5-r2 Has Been Released peterleko [2004/04/23-02:39] |
| Response : Edit : Delete |
|
Is
that because UTS has a "soft" carpet. |
| V 1056 ERS-7 second tilt joint PID value ejt [2004/04/21-11:38] |
| Response : HomePage : Email : Edit : Delete |
|
Just
a suggestion - it seems the second tilt joint's 'I' value for the PID parameters is perhaps too low... at least for my taste. The joint can be over .075 radians (~4.3 degrees) from the commanded value, but not move, probably due to static friction within the joint balancing the motor. By increasing the 'I' parameter from 0x04 to at least 0x10, the joint will keep a tighter bound on the commanded value, without any oscillation. (far from it - it can still take a few seconds to drift into center view, but at least it will do so) Normally this is a small error, but since the camera is being affected, this can make a noticable difference between an object being centered in the image or not. -ethan |
| V 1034 My aibo(ERS-7) cannot run open-r samples. help !! aiboKang [2004/04/05-23:10] |
| | Response : Edit : Delete |
|
|
what a bad problem ! My aibo(ers-7) can't run open-r samples! | This happened when I tried to combine two samples ( ImageObserver and | MoNetTest) into one (ImageCapture) last week. My aibo worked with open-r | samples (such as HelloWorld, BlinkingLED7, BallTrackingHead7) perfectly at | first, and everything was ok until I tried to combine ImageObserver and | MoNetTest into ImageCapture. I combined ImageObserver and MoNetTest into | ImageCapture, and run it on aibo but failed. Aibo didn't do anything expected. | Then I took the "PMS " out of aibo and re-compiled my ImageCapture program on | the computer. I found some mistakes in the program, which was wrong locators | of sensors. After correcting it and compiling successfully, I re-run the | ImageCapture on aibo but failed again! When I put pause button down, my aibo | did nothing with pause button indicator green, and after about 8 seconds | a voice was heard as if a motor moved. Moreover , it took about as long as 3 | minutes to shut down aibo! | To ensure aibo properly working with open-r, I re-run | sample "HelloWorld" which run correctly before. To my surprise, | the "HelloWorld" sample cannot work! Actually not only " HelloWorld" but also | BlinkingLED7 and BallTrackingHead7, which run properly before, failed to | work. And more surprisingly, all of them had the same malfunction,and the | same feedback information from aibo through wireless LAN. The feedback | information indicated initialization information as follows: | ....... (omitted ) | | No Name Context OID | 1 systemCore 0x80280c20 0xffffffff | 2 ...... | ...... ( omitted ) | | 58 hookConsoleIOAct 0x80292b60 0x800000043 | _ ( stopped here ) | | To know if aibo was all right, I inserted the original MS into aibo, and | found it as well as formerly. Now, we can know that the aibo , "PMS" , and | open-r samples are all right respectively, so I guess the problem seemingly | happened to the basic objects provided with the system, right? | Is it because of the wrong locators of sensors I set in my " ImageCapture" | program which resulted in this or anything else ? Do I have to send my aibo | to Sony for warranty? | | Who can help me ? Thanks so much! | I'm really in hot waters!! | | (The article that 2004/04/05-23:04 posted, was modified by Author) | |
| +- 1055 RE:My aibo(ERS-7) cannot run open-r samples. help !! openrsupport@(Administrator) [2004/04/21-00:40] |
| Response : Edit : Delete |
|
It
appears that your objects are not being loaded. Make sure that your
objects are listed in /OPEN-R/MW/CONF/OBJECT.CFG. For the HelloWorld sample, make sure that /OPEN-R/MW/OBJS/HELLO.BIN and /OPEN-R/MW/OBJS/POWERMON.BIN have also been copied to your Memory Stick. If you continue to have problems, delete all files on your Memory Stick, copy one of the original Memory Stick images to your Memory Stick, and try again. OPEN-R SDK Support |
| V 1051 Error installing 1.1.5 on Linux malcolmr [2004/04/16-15:47] |
| | Response : Email : Edit : Delete |
|
| I am
having trouble installing version 1.1.5 of the SDK on my linux machine.
I | have been following the instructions set out at: | | http://www-2.cs.cmu.edu/~tekkotsu/SDKInstall.html | | but I'm getting compilation errors when I run the build-devtools-3.3.2-r1.sh script: | | ../../binutils-2.14/bfd/archive.c:1823: error: `PACKAGE' undeclared (first use | in this function) | | It appears that "archive.c" includes "sysdep.h" which in turn includes | "config.h" and the compiler is finding "intl/config.h" before "bfd/config.h". | The latter file contains a definition of PACKAGE. The former one does not. If I | remove "../intl/" from the INCLUDES variable in the Makefile it compiles without | any problems. Similar problems appear in other directories. | | I am using gcc 3.3.2 on Debian Linux machine (I don't know which release. I | don't maintain this machine.) | | Malcolm | | | |
| +- 1053 RE:Error installing 1.1.5 on Linux malcolmr [2004/04/19-14:18] |
| Response : Email : Edit : Delete |
|
I
found the problem: I had "." in my C_INCLUDE_PATH. gcc ignores
-I<dir> if <dir> is already in the C_INCLUDE_PATH. This means that -I. was being ignored, and so -I../intl was taking precedence. I fixed this, and binutils now compiles fine, but I'm getting another bug while compiling gcc. While running "make" in bld-gcc-3.3.2/gcc/ I get the following: cc1: unrecognized option `-auxbase' cc1: output filename specified twice make: *** [g++spec.o] Error 1 Malcolm |
| V 1041 More problems with BATTERY_OVER_CURRENT omega697 [2004/04/07-05:53] |
| | Response : Email : Edit : Delete |
|
| We
Austin Villa folks are having a problem with the BATTERY_OVER_CURRENT
crash. | Not really a crash, per se, but a controlled shutdown which results from too | much current being drawn from the battery (or something like that). We changed | our joint gains to the weak gains hoping it would fix the problem, but | unfortunately, it has resurfaced. The problem seems to crop up when the robots | are either changing direction really quickly or starting to move. We're not | really sure what to do about this. If anyone else is still experiencing this, | we'd like to work with you to solve it. I doubt this is the kind of research | challenge the RoboCup folks have in mind for us to compete over... | |
| +- 1044 RE:More problems with BATTERY_OVER_CURRENT Sauber [2004/04/08-19:02] |
| | Response : Email : Edit : Delete |
|
| We
are still getting the occasional BATTERY_OVER_CURRENT shutdown. I agree
that | it appears to happen when changing direction quickly (or for that matter change | the direction of a motor very fast). We have had shutdowns during regular | (smooth) motions as well, however. | | I believe we can monitor the current (current is reported in the OPowerStatus | struct) but it would be helpful if Sony could give us some further information. | For example, what formula is used to generate the shutdown? | | If we knew more details we may be able to monitor the current and deal with the | problem before the AIBO shuts off. | | -NUbots (University of Newcastle) | |
| +- 1049 RE:More problems with BATTERY_OVER_CURRENT openrsupport@(Administrator) [2004/04/13-01:56] |
| Response : Edit : Delete |
|
The
high-current detection threshold has been changed in OPEN-R SDK 1.1.5
r2. Please use this updated version of the OPEN-R SDK. If you continue to have problems with BATTERY_OVER_CURRENT shutdowns, please inform us on this BBS. Thank you, OPEN-R SDK Support |
| V 1043 Download Open-R SDK kingdonk [2004/04/07-21:04] |
| | Response : Email : Edit : Delete |
|
| I am
trying to download the Open-R SDK but i always get a message saying that | the sever can not be found. | ANy way around this or others with the same problem. | |
| |- 1045 RE:Download Open-R SDK TodboT [2004/04/09-00:47] |
| | Response : Edit : Delete |
|
| I
haven't had any problems. I think I remember somebody saying to clear
your | Internet cache and to try again. | |
| +- 1047 RE:Download Open-R SDK openrsupport@(Administrator) [2004/04/09-09:37] |
| Response : Edit : Delete |
|
>I
am trying to download the Open-R SDK but i always get a message saying
that >the sever can not be found. >ANy way around this or others with the same problem. Hi, Which file are you attempting to download? Thanks, OPEN-R SDK Support |
| V 1040 how to get force information from joints? fidelwahoo [2004/04/07-05:01] |
| | Response : Edit : Delete |
|
| Does
anyone know how i could go about getting force data sensed at the
joints, | or if this is even possible? | | I see a structure called OForce in the Level 2 Reference Guide, but it is | unclear what sort of sensors this information would come from. I'm hoping it | can be obtained for the joints, and i saw something on the Tekkotsu webpage that | seemed to indicate that ("18 PID joints, each with force sensing"), but the | OSensorFrameVectorData i get from OVirtualRobotComm seems to only be giving me | OJointValues. | | Any help would be greatly appreciated.... | |
| +- 1046 RE:how to get force information from joints? ejt [2004/04/09-03:05] |
| Response : Edit : Delete |
|
>
i saw something on the Tekkotsu webpage that >seemed to indicate that ("18 PID joints, each with force sensing") If you're using Tekkotsu, you can just access the state->pidduties[] array directly. This is kept up to date by the framework in WorldState::::read(OSensorFrameVectorData& ...) http://www-2.cs.cmu.edu/~tekkotsu/dox/WorldState_8cc-source.html#l00071 In particular, you'd want to see the GETDUTY macro used in that file to see how to get the duty (aka force) information: http://www-2.cs.cmu.edu/~tekkotsu/dox/WorldState_8cc-source.html#l00021 Remember though, the duty cycle information is a measurement of how hard the motor is working - this could be due to fighting the inertia of the joint simply because the joint is moving, or could be due to an outside influence because you're pressing against something... it can be tricky to tell which is which. |
| V 1032 Author DELETE ***** [2004/04/08-20:32] |
| | Response : Edit : Delete |
|
| (The
article that 2004/04/04-03:04 posted, was modified by Author) | |
| +- 1042 Author DELETE ***** [2004/04/08-19:07] |
| Response : Edit : Delete |
|
(The
article that 2004/04/07-13:51 posted, was modified by Author) |
| V 1039 I get the mkbin: can't analyze the result of readelf error outsider [2004/04/07-03:03] |
| Response : Edit : Delete |
|
Hi, I've installed the OPEn-R SDK under a Debian Linux. I've also installed with no problem the compilation utils (gcc, binutils and newlib) using the provided script. Nevertheless, when I try to compile ANY program, I always get the following error message: mkbin: can't analyze the result of readelf error Can anybody please tell me how can I solve this problem. Domo arigato. |
| V 1036 Resurrect AiboMind memory stick ddlee [2004/04/06-11:59] |
| | Response : HomePage : Email : Edit : Delete |
|
| I thought I would try using the Aibo Mind memory stick that | comes with the ERS-7 to run an OPEN-R SDK program. | I first copied all the files from Aibo Mind's OPEN-R directory, | and then installed the OPEN-R SDK files and my compiled program. | No luck--robot doesn't even get past the initial system boot. | So then I copied the Aibo Mind files back to the memory stick. | Again no luck--robot gives back the unhappy musical tones | and turns off! | Does anyone know if there is anything I can do with this memory | stick now? Thanks. | -Dan | |
| +- 1038 RE:Resurrect AiboMind memory stick AiboPet [2004/04/07-02:21] |
| Response : Edit : Delete |
|
The first two questions: Are you using a SONY brand stick reader ? Did you format the memory stick ? ============================== >I thought I would try using the Aibo Mind memory stick that >comes with the ERS-7 to run an OPEN-R SDK program.... If you are using a Sony brand stick reader and have not formatted the memory stick, then what you have done should not cause any permanent harm to the stick (other than wasting your time) >I first copied all the files from Aibo Mind's OPEN-R directory, >and then installed the OPEN-R SDK files and my compiled program. As pointed out several times, this will not work. You need a Programmable Memory Stick (PMS AKA R-CODE stick), one without an ERF- number on it (AiboMind has an ERF- number) >So then I copied the Aibo Mind files back to the memory stick. Possibility #1 You didn't copy the files back properly, and the stick is ok. Try downloading a 'virgin' image from the aibo-friends.org BBS and unzip it to the stick. Delete the old files first, do not format. Possibility #2: The stick's copy protection is broken (because of using a non-Sony brand reader or because of full formatting the stick). In this case you can try to get a replacement from Sony. Otherwise you have turned your 32MB pink stick into a 32MB purple stick (ie. can be used in cameras and PDAs, but not AIBO) |
| V 1037 game controller problems ckch [2004/04/07-00:19] |
| Response : HomePage : Email : Edit : Delete |
|
Hello, Does anyone know how to set up flow control in the latest game-controller with 1.1.5 SDK. At the moment, it seems to have no effect when changing the baud rate limit. Also, when running the latest game-controller, if any dog(s) configured is not connected, then closing the controller will lead to a problem where the next time u open the controller, it says "there is not enough free disk space", and i have to reboot my machine in order to reopen the controller. These problems are quite annoying and serious. Does anyone know how to get around them :-} |
| V 1030 timers and sleep functionallity in OPENR aibourjc [2004/04/02-00:52] |
| | Response : Email : Edit : Delete |
|
| I am
looking for information about the use of timers, and about something
that | provides me something like sleep functionality, but I cannot find anything. Has | any of you use this? | Thanks you in advance | |
| +- 1035 RE:timers and sleep functionallity in OPENR TodboT [2004/04/06-08:05] |
| Response : Edit : Delete |
|
Something like this should work: longword GetTime() { SystemTime cur; GetSystemTime(&cur); return cur.seconds * 1000 + cur.useconds/1000; } |
| V 1031 packet loss when pinging ERS-7 NWCIL [2004/04/02-07:11] |
| | Response : Email : Edit : Delete |
|
| Hi everyone - | I was excited to finally ping my ERS-7 for the first time today, but dismayed | to find that it doesn't ping reliably. In a set of 100 pings, I typically find | 80% packet loss. This is while running the simple "Hello World" OPEN-R | program. I've only been able to telnet to it successfully once. | | Any ideas?? | thanks, | Bryan | |
| +- 1033 RE:packet loss when pinging ERS-7 ejt [2004/04/05-12:15] |
| Response : HomePage : Email : Edit : Delete |
|
check
your wireless connection strength - this is typical of what happens
when the wireless signal strength is poor. We had to set up a private access point in our lab because we didn't get good enough reception on the campus wireless network (although reception is pretty mediocre to begin with in our lab, so that's not really a surprise) In general, AIBOs don't tend to get as good reception as some laptops. Or at least the 210 didn't. I should see how far our ERS-7 can stroll... maybe now i can make it down to the vending machines and back ;) Take a look at the 'ERA201D1Info' sample (in 'common' of the sample code download) This can display some wireless network stats for you (i'm rusty on what exactly it gives though) -ethan |
| V 1027 How is the 'OLEDCommandValue3' defined? aiboKang [2004/03/29-18:07] |
| | Response : Edit : Delete |
|
| Hello,anybody, | Here is a problem. | When I run 'BlinkingLED7' sample , I found the 'OLEDCommandValue3' was used | as the data type of LED command , which is 'OLEDCommandValue2' in the PDF file | (Level2ReferenceGuide.pdf) provided by Sony . Actually there is nothing | about the description of 'OLEDCommandValue3' in the PDF file | (Level2ReferenceGuide.pdf) . Is it missed in the PDF file or something else? | what on earth is 'OLEDCommandValue3' like ? | | Can anybody tell? Thanks in advance. | |
| +- 1028 RE:How is the 'OLEDCommandValue3' defined? AiboPet [2004/03/30-00:56] |
| | Response : Edit : Delete |
|
| >When
I run 'BlinkingLED7' sample , I found the 'OLEDCommandValue3' was used | >as the data type of LED command ... | | Download the latest documentation and check the ModelInformation_7_E.pdf file. | | >what on earth is 'OLEDCommandValue3' like ? | | The ERS-7 face LEDs are tricky. You can optionally set the intensity and the | mode (MODE A or B, which make the LEDs light up in different ways). Not all | combinations work. | | Check out the header "ODataFormats.h" for the structure and the PDF for | explaining which modes work (it has little diagrams to show which LEDs are | which) | |
| +- 1029 RE:How is the 'OLEDCommandValue3' defined? aiboKang [2004/03/30-11:29] |
| Response : Edit : Delete |
|
>>When
I run 'BlinkingLED7' sample , I found the 'OLEDCommandValue3' was used >>as the data type of LED command ... > >Download the latest documentation and check the ModelInformation_7_E.pdf file. > >>what on earth is 'OLEDCommandValue3' like ? > >The ERS-7 face LEDs are tricky. You can optionally set the intensity and the >mode (MODE A or B, which make the LEDs light up in different ways). Not all >combinations work. > >Check out the header "ODataFormats.h" for the structure and the PDF for >explaining which modes work (it has little diagrams to show which LEDs are >which) I got it! It's very kind of you! |
| V 1025 What is different between OPEN-R SDK 1.1.5 and oldr version? fjuee922 [2004/03/26-22:44] |
| | Response : Email : Edit : Delete |
|
| What is different between OPEN-R SDK 1.1.5 and oldr version? | |
| +- 1026 What is different between OPEN-R SDK 1.1.5 and oldr version? fjuee922 [2004/03/26-22:47] |
| Response : Email : Edit : Delete |
|
>What
is different between OPEN-R SDK 1.1.5 and oldr version? and cypwin, gcc all update to V1.1.5 What's the different? |
| V 1018 Problems with the new OPENR-SDK 1.1.5.r1 angel [2004/03/23-22:34] |
| | Response : HomePage : Email : Edit : Delete |
|
| Recently
I have upgraded to the newly released OPEN-R 1.1.5-r1 | Since then I have been unable to get running any of the programs for Aibo. | I have tried to recompile HelloWorld and install it on the memory stick. Aibo, | according to my wireless HUB, does connect; however I am not able to either | | ping 192.168.2.11 | | or | | telnet 192.168.2.11 59000 | | I have tried with static IP and setting up DHCP on the wireless. Nothing appears | to work. Here is a copy of my WLANCONF.TXT | | # | # WLAN | # | HOSTNAME=snoopy | ESSID=solar | WEPENABLE=1 | WEPKEY=0x03142368728354226713744267 | APMODE=1 | | # | # IP network | # | #USE_DHCP=1 | # | # If DHCP is not used (USE_DHCP=0), you need to specify IP | # network configuration. | # | USE_DHCP=0 | ETHER_IP=192.168.2.11 | ETHER_NETMASK=255.255.255.0 | IP_GATEWAY=192.168.2.1 | DNS_SERVER_1=192.168.2.1 | | # | # SSDP | # | SSDP_ENABLE=1 | | Now I know that this file works because it is the same one install on the | Aibomind stick, and that still works. | | I have tried doing a complete re-installation of the SDK complete from scratch. | Still nothing works on Aibo. | | Another interesting thing is that when I run ObjectComm on the host PC using | R-OPENR then it does work as expected. So it would seem that the compiled | objects appear to work on the host as expected. I supposed the problem is someho | w wireless related but I am not sure why it would work with AiboMind and not | with HelloWorld. | | I have made sure that I follow the installation guide step by step. I left the | original WLANDFLT.TXT on the stick and added my WLANCONF.TXT there. | | Any suggestions as to how to figure the problem? | | Thanks | |
| +- 1020 RE:Problems with the new OPENR-SDK 1.1.5.r1 AiboPet [2004/03/24-00:06] |
| | Response : Edit : Delete |
|
| >Recently
I have upgraded to the newly released OPEN-R 1.1.5-r1 | >Since then I have been unable to get running any of the programs for Aibo. | ... | > Now I know that this file works because it is the same one install on the | > Aibomind stick, and that still works. | | AiboMind uses a different LAN component. | Is this the first time you tried the OpenR SDK LAN connection ? | | --- | The following features did not work on the old OpenR SDK LAN, so they may be | the problem. Take a look at the WLANDFLT.TXT file. | (I'm not sure what if any have been added recently, but if you trim this back | to safer values, that should help) | | >WEPKEY=0x03142368728354226713744267 | -- only 5 character ASCII WEP strings | >APMODE=1 | -- I recommend people use APMODE=2 (auto) | | >USE_DHCP=0 | -- this is not required, remove it to be safe | >DNS_SERVER_1=192.168.2.1 | -- this is not required, remove it to be safe | >SSDP_ENABLE=1 | -- this is not supported, remove it to be safe | | =========== | >Another interesting thing is that when I run ObjectComm on the host PC using | >R-OPENR then it does work as expected. So it would seem that the compiled | >objects appear to work on the host as expected. I supposed the problem is | someho | >w wireless related | >... but I am not sure why it would work with AiboMind and not | >with HelloWorld. | | AiboMind and OpenR SDK programs (like HelloWorld) should be thought of as two | different worlds. | |
| +- 1021 Problem with SDK 1.1.5.r1 angel [2004/03/24-00:15] |
| | Response : HomePage : Email : Edit : Delete |
|
| >AiboMind
uses a different LAN component | >Is this the first time you tried the OpenR SDK LAN connection ? | | I have used a similar configuration file with beta I think it was and | everything worked smoothly, including my hexadecimal WEP According to the | INSTALLATIONpdf hexadecimal is supported, so unless it is broken now I am not | sure how to explain the fact that it worked for my network before Anyhow I have | tried disabling WEP too and that didn&#;t help I will do more tests removing | lines from the WLAN config file | | > | > | >The following features did not work on the old OpenR SDK LAN, so they may be | >the problem Take a look at the WLANDFLTTXT file | >I&#;m not sure what if any have been added recently, but if you trim this back | >to safer values, that should help | > | >>WEPKEY=x | > only character ASCII WEP strings | This I truly don&#;t understand I did work for me on beta, | so unless it is not supported now I can&#;t explain how the robot communicated | with my encripted robot before | | >>APMODE= | > I recommend people use APMODE= auto | > | >>USE_DHCP= | > this is not required, remove it to be safe | >>DNS_SERVER_= | > this is not required, remove it to be safe | >>SSDP_ENABLE= | > this is not supported, remove it to be safe | | I will remove those lines and try again | | >=========== | | >AiboMind and OpenR SDK programs like HelloWorld should be thought of as two | >different worlds | | Good to know | (The article that 2004/03/24-00:15 posted, was modified by Author) | |
| +- 1023 Problem with rel 1.1.5.r1 angel [2004/03/24-00:59] |
| | Response : HomePage : Email : Edit : Delete |
|
| Nothing
works | | >>AiboMind uses a different LAN component | >>Is this the first time you tried the OpenR SDK LAN connection ? | > | >I have used a similar configuration file with beta I think it was and | >everything worked smoothly, including my hexadecimal WEP According to the | >INSTALLATIONpdf hexadecimal is supported, so unless it is broken now I am not | >sure how to explain the fact that it worked for my network before Anyhow I have | >tried disabling WEP too and that didn&#;t help I will do more tests removing | >lines from the WLAN config file | > | >> | >> | >>The following features did not work on the old OpenR SDK LAN, so they may be | >>the problem Take a look at the WLANDFLTTXT file | >>I&#;m not sure what if any have been added recently, but if you trim this back | >>to safer values, that should help | >> | >>>WEPKEY=x | >> only character ASCII WEP strings | >This I truly don&#;t understand I did work for me on beta, | >so unless it is not supported now I can&#;t explain how the robot communicated | >with my encripted robot before | > | >>>APMODE= | >> I recommend people use APMODE= auto | >> | >>>USE_DHCP= | >> this is not required, remove it to be safe | >>>DNS_SERVER_= | >> this is not required, remove it to be safe | >>>SSDP_ENABLE= | >> this is not supported, remove it to be safe | > | >I will remove those lines and try again | > | >>=========== | > | >>AiboMind and OpenR SDK programs like HelloWorld should be thought of as two | >>different worlds | > | >Good to know | >The article that posted, was modified by Author | (The article that 2004/03/24-00:59 posted, was modified by Author) | |
| +- 1024 Update 1.1.5 Beta1 works but..... angel [2004/03/24-02:09] |
| Response : HomePage : Email : Edit : Delete |
|
I
went back to SDK 1.1.5beta1 and after detecting the same error with
HelloWorld (no ping response) and immediately assume that the wireless HUB had a corrupted rounting table. After reseting it and eliminating WEP, enabling DHCP, and with a very small WLANCONF.txt file, now HelloWorld works again. Next I am going to try with SDK 1.1.5 rel 1 to see if my problem was only in the wireless router. |
| V 1019 More problems with OPEN_R_SDK-1.1.5.r1 angel [2004/03/24-00:06] |
| | Response : HomePage : Email : Edit : Delete |
|
| Has
anyone succeded at installing the linux version of OPENR 1.1.5.r1? | I have redhat 9 and I am having lots of troubles with the build-devtools-3.3.2-r1.sh | | It seems as if some directories are missing in the gcc library. I am in the | process of narrowing down the error and generating a log file. I will post it soon. | |
| +- 1022 Errors installing 1.1.5.r1 on Linux angel [2004/03/24-00:31] |
| Response : HomePage : Email : Edit : Delete |
|
>Has
anyone succeded at installing the linux version of OPENR 1.1.5.r1? >I have redhat 9 and I am having lots of troubles with the build-devtools-3.3.2-r1.sh > >It seems as if some directories are missing in the gcc library. I am in the >process of narrowing down the error and generating a log file. I will post it soon. Here is the errors I got when running the build script: Unpacking binutils-2.14.tar.gz Unpacking gcc-3.3.2.tar.gz tar: gcc-3.3.2/libjava/java/security/cert/CertificateFactorySpi.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/cert/CertificateParsingException.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/cert/X509CRLEntry.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/cert/X509Extension.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/AccessControlException.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/AlgorithmParameterGenerator.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/AlgorithmParametersSpi.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/BasicPermission.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/CodeSource.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/DigestOutputStream.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/DummyMessageDigest.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/GeneralSecurityException.java: Wrote only 0 of 10240 bytes tar: gcc-3.3.2/libjava/java/security/GuardedObject.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/IdentityScope.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/InvalidParameterException.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/KeyException.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/KeyFactorySpi.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/KeyPairGenerator.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/KeyStore.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/MessageDigest.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/NoSuchAlgorithmException.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/Permission.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/Policy.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/PrivilegedAction.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/PrivilegedExceptionAction.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/Provider.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/SecureRandom.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/SecurityPermission.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/SignatureException.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/SignedObject.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/UnresolvedPermission.java: Wrote only 0 of 10240 bytes tar: Skipping to next header tar: gcc-3.3.2/libjava/java/security/interfaces: Cannot mkdir: No space left on device tar: gcc-3.3.2/libjava/java/security/interfaces/DSAKey.java: Cannot open: No such file or directory tar: gcc-3.3.2/libjava/java/security/interfaces/DSAKeyPairGenerator.java: Cannot open: No such file or directory tar: gcc-3.3.2/libjava/java/security/interfaces/DSAParams.java: Cannot open: No such file or directory ta |
| V 1015 Commanding AIBO wirelessly kabossa [2004/03/16-05:51] |
| | Response : Email : Edit : Delete |
|
| Can
someone please help me figure out how to make aibo make a sound. | | I need a program on a remote computer (not AIBO) to issue a command to the | robot which will make Aibo bark. | | ie | | Aibo is turned on and doing whatever | | computer (not aibo) sends a command to aibo (wirelessly) | aibo receives this command and barks (or plays any sound) | | if anyone can help me do this, I would GREATLY appreciate it. | | Thanks!! | |
| |- 1016 RE:Commanding AIBO wirelessly ejt [2004/03/16-11:39] |
| | Response : HomePage : Email : Edit : Delete |
|
| my quick answer: Load up Tekkotsu ( http://www.tekkotsu.org/ ), select the File Access menu, | then Play Sound, then pick a file to play (reads the file listing directly from the memory stick - just | copy over your own file if you don't like one of the included barks) | | If you're picky, you can customize various ways - e.g., turn off controller sounds in ms/config/ | tekkotsu.cfg, or write some custom code to play various sounds based on different inputs. | | PS new users planning to develop code should check out the current CVS version if they're also | using the recently released SDK. | |
| +- 1017 RE:Commanding AIBO wirelessly AiboPet [2004/03/17-01:48] |
| Response : Edit : Delete |
|
>Can
someone please help me figure out how to make aibo make a sound. > >I need a program on a remote computer (not AIBO) to issue a command to the >robot which will make Aibo bark. Another alternative for simple cases like this is to use RCODE/RCodePlus. (currently only for the ERS-210/220, but RCode for the ERS-7 has been hinted as coming soon). You can control AIBO using telnet strings over the WiFi connection. This works best if you have a canned set of motions and sounds that you can store on the memory stick, and then use simple telnet commands to make AIBO do things. eg: "PLAY ACTION O2_CUTE_BARK2" (a canned barking sound) or "PLAY ACTION WALK 0 1000" (walk a metre) You can also dynamically upload sounds and motions to the dog for performing, but that is more complicated |
| V 1014 Broken Aibo? akoster [2004/03/10-22:23] |
| Response : HomePage : Edit : Delete |
|
Hi, We have an ERS-7 with a complete joint lock. The right hind leg can't turn anymore at all. We tried resetting the robot and removing the battery completely, but to no avail, the joint stays completely jammed. Before unscrewing it or sending it back to Sony for Warranty we were wondering whether anybody has had this same problem and got a solution? Kind regards, Andrew Koster Dutch Aibo Team Utrecht University |
| V 1013 Important Attention Updated openrsupport@(Administrator) [2004/03/09-03:16] |
| Response : Edit : Delete |
|
Important
attention when using the OPEN-R SDK has been updated. http://openr.aibo.com/openr/eng/no_perm/notice.html Thank you, OPEN-R SDK Support |
| V 1006 I can not install OPEN-R-SDK in Linux. brightring [2004/03/04-09:57] |
| | Response : Email : Edit : Delete |
|
| My
OS is RedHat Linux 7.3. I place devtools-3.2-r1.sh,
binutils-2.14.tar.gz, | gcc-3.3.2.tar.gz and newlib-1.10.10.tar.gz in the same folder, | then execute devtools-3.2-r1.sh ( with root user). But the result is: | | Unpacking binutils-2.14.tar.gz | tar: binutils-2.14/binutils/doc/Makefile.am: Cannot change ownership to uid | 1000, gid 1000: Operation not permitted | tar: binutils-2.14/binutils/doc/Makefile.in: Cannot change ownership to uid | 1000, gid 1000: Operation not permitted | ...... | | What is the problem and how to solve it? Thank you | (The article that 2004/03/04-09:52 posted, was modified by Author) | |
| +- 1012 RE:I can not install OPEN-R-SDK in Linux. brightring [2004/03/08-11:12] |
| Response : Email : Edit : Delete |
|
I've solved the problem myself! The files should be placed in the Linux partition! |
| V 1008 210 R-Code sharkboy717 [2004/03/06-08:30] |
| | Response : Edit : Delete |
|
| Does
anyone know how long it took for the original R-Code for the 210 to be | released? | |
| +- 1010 RE:210 R-Code AiboPet [2004/03/07-03:08] |
| | Response : Edit : Delete |
|
| >Does
anyone know how long it took for the original R-Code for the 210 to be | >released? | | (RCODE history lesson) | | The original ERS-210 robots were available in late 2000, right before Xmas. | | The first RCODE came out in Japan in Feb 2001, along with the WiFi LAN cards, | PMS memory sticks and Master Studio. | It took until late April / early May 2001 for the English version to come out. | From what I understand localizing the Master Studio software was the | bottleneck. There was a minor update to RCODE at that time (RCODE 1.1, Master | Studio 1.01). | | The only major upgrade to RCODE came out in Feb 2002 (RCODE 1.2, Master Studio | 1.1). They added support for the 31x and 220 models as well as other minor bug | fixes. | |
| +- 1011 RE:210 R-Code sharkboy717 [2004/03/07-03:11] |
| Response : Edit : Delete |
|
Thx
AiboPet! I hope r-code for the 7 will follow that schedule and we'll
have an english version by early may! dunno if my hopes are justified though :-( |
| V 1009 Floating Point Exception ejt [2004/03/06-10:36] |
| Response : HomePage : Email : Edit : Delete |
|
Just
curious if there was something that could be done to *not* crash on a floating point exception? It's not always such a bad thing... Specifically, I'm dealing with some rotations, where a rotation of infinite radius is simply a translation. It makes the code very straightforward if I could do a division: radius=(distance/2)/sin(arc_angle/2) and then use isinf() or finite() later to test for the translation case (arc_angle==0). This works on a desktop, where it returns an 'inf' float instead of just dying. Just a thought... thanks -ethan |
| V 0970 ers-7 crashes with dive kick dvail2 [2004/02/13-07:05] |
| | Response : Email : Edit : Delete |
|
| Hi, | | We're having a problem where our ERS-7 robots crash if they perform a dive kick | onto the orange ball for RoboCup. The robots run normally when not jumping onto | the ball. They do not always crash when diving - they crash roughly 20% of the | time when they dive. | | We've discovered that the DoDestroy method of our object is being called from | somewhere else on the robot, apparently in response to the sudden jolt. Has | anyone else encountered this? Is there anyway to turn it off? Since our code is | running in the context of RoboCup, we don't mind jolting the robot even if it | does risk damaging it. We'd like to stop DoDestroy from being called if possible. | | thanks in advance, | -Doug | |
| |- 0980 RE:ers-7 crashes with dive kick GTleader [2004/02/18-23:40] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | Our
robots crash in nearly any situation. Some of them are reproducible,
such | | as or old ERS-210 "grabBall", and sometimes the robot crashes for no appearent | | reason. In many cases, no emon.log is written, in some it is, stating that the | | robot crashed when calling NumberOfObservers() (at least in the code of the | | GermanTeam) from an arbitrary module. I iterpret this as a side effect of | | another module that already died. At least in some cases, a message | | about "JAM" is written to the serial port. However, since we rarely have a | | serial connection to the robots, we don't know whether this is always the case. | | | | If this problem is not solved, we will see many dead ERS-7 in the games of the | | next RoboCup. | | | | Thomas | | |
| | |- 0982 RE:ers-7 crashes with dive kick ejt [2004/02/19-04:04] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | >At
least in some cases, a message | | >about "JAM" is written to the serial port. However, since we rarely have a | | >serial connection to the robots, we don't know whether this is always the case. | | Don't forget if you use the WCONSOLE system binaries, you can telnet to port | | 59000 over wireless while the Aibo(s) is/are running. For system errors that | | cause a shutdown, you will then consistently get the error message that caused | | it, before it shuts down. | | It's of more limited use for general debugging because a crash will often cause | | any network traffic in buffers to be lost, so the last couple KB may not be | | transmitted. But the system errors aren't "crashes" per se, since it does a | | controlled shutdown, so the buffers will be cleared. | | | | On a related note, has anyone gotten blocking network transmission to work? Our | | attempt at it reduces the amount of buffering that is lost on a crash, but the | | block seems to return before the data has actually left the Aibo. (see the | | CrashTestBehavior in Tekkotsu for a demo) | | | | -ethan | | |
| | +- 1005 RE:ers-7 crashes with dive kick omega697 [2004/03/03-17:36] |
| | Response : Email : Edit : Delete |
|
| We
had that problem for quite a while on our ERS7 robots too. We changed
the | joint gains from "normal" to "weak" and the crash stopped. Most of the time we | got a "BATTERY_OVER_CURRENT" message on the telnet screen, but no emon.log. | | -Kurt | | >Our robots crash in nearly any situation. Some of them are reproducible, such | >as or old ERS-210 "grabBall", and sometimes the robot crashes for no appearent | >reason. In many cases, no emon.log is written, in some it is, stating that the | >robot crashed when calling NumberOfObservers() (at least in the code of the | >GermanTeam) from an arbitrary module. I iterpret this as a side effect of | >another module that already died. At least in some cases, a message | >about "JAM" is written to the serial port. However, since we rarely have a | >serial connection to the robots, we don't know whether this is always the case. | > | >If this problem is not solved, we will see many dead ERS-7 in the games of the | >next RoboCup. | > | >Thomas | |
| |- 0998 RE:ers-7 crashes with dive kick openrsupport@(Administrator) [2004/03/02-07:57] |
| | Response : Edit : Delete |
|
| Hi, | | The newest OPEN-R SDK (1.1.5) contains a feature called | "Jam Detection High Threshold". If you are finding | that the jam detection routine is too sensitive, | you can make it less strict by creating the file | "/OPEN-R/SYSTEM/CONF/VRCOMM.CFG" and writing | a single line, "JamDetectionHighThreshold" in the file. | | This only works with ERS-7. | | Thanks, | | OPEN-R SDK Support | |
| +- 1004 RE:ers-7 crashes with dive kick omega697 [2004/03/03-17:34] |
| Response : Email : Edit : Delete |
|
We
had a similar problem last year with the ERS-210. The problem was that
when the AIBO hit the ground, the battery became dislodged from the robot just enough that it powered down the robot. By stuffing some cardboard under the battery case, we were able to prevent the problem. -Kurt >Hi, > >We're having a problem where our ERS-7 robots crash if they perform a dive kick >onto the orange ball for RoboCup. The robots run normally when not jumping onto >the ball. They do not always crash when diving - they crash roughly 20% of the >time when they dive. > >We've discovered that the DoDestroy method of our object is being called from >somewhere else on the robot, apparently in response to the sudden jolt. Has >anyone else encountered this? Is there anyway to turn it off? Since our code is >running in the context of RoboCup, we don't mind jolting the robot even if it >does risk damaging it. We'd like to stop DoDestroy from being called if possible. > >thanks in advance, >-Doug |
| V 0997 The W3AIBO sample sharkboy717 [2004/03/02-05:52] |
| | Response : Edit : Delete |
|
| I've
been able to set up the W3AIBO example included with the SDK, but i
want to | be able to move my 7's head around remotely. Does anybody know how i would be | able to do this? (sry if it sounds stupid, but i've barely done any C++ | programming!) | |
| +- 1001 RE:The W3AIBO sample TodboT [2004/03/03-10:53] |
| | Response : Edit : Delete |
|
| You
should be able to combine the W3AIBO with EchoServer and MovingHead7. | W3AIBO can stay the same, but you'll have to modify EchoServer and MovingHead7 | to use inter-object communication (ObjectComm). EchoServer would receive | commands remotely, and pass a message to MovingHead7. MovingHead7 will have to | be modified to read the message from EchoServer and be able to move the head to | a desired position (instead of just panning left and right). Good luck! | |
| +- 1002 RE:The W3AIBO sample sharkboy717 [2004/03/03-11:09] |
| | Response : Edit : Delete |
|
| thanks!
just a ques: i haven't really done that much c++ before, is there any | really good way to learn about how i would do what you described below (other | than by closing my eyes and pounding the keys *_*)? | | >You should be able to combine the W3AIBO with EchoServer and MovingHead7. | >W3AIBO can stay the same, but you'll have to modify EchoServer and MovingHead7 | >to use inter-object communication (ObjectComm). EchoServer would receive | >commands remotely, and pass a message to MovingHead7. MovingHead7 will have to | >be modified to read the message from EchoServer and be able to move the head to | >a desired position (instead of just panning left and right). Good luck! | |
| +- 1003 RE:The W3AIBO sample TodboT [2004/03/03-11:49] |
| Response : Edit : Delete |
|
There's
a pretty good tutorial here: http://www.ensta.fr/~baillie/openr_tutorial.html Specifically for what you want to do; start with the sample programs, and make small minor changes, eventually massaging them into what you want them to do. You'll need to understand how inter-object communication works (ObjectComm). Try adding this feature to EchoServer and MovingHead7. Once they can pass messages, change MovingHead7 so it moves to a target instead of side to side. Once those are done, it should be simple to tie everything together. Read the tutorial, manuals, and try to understand how the samples work. I hope this helps. |
| V 1000 OPEN-R SDK 1.1.5 documentation released openrsupport@(Administrator) [2004/03/03-09:07] |
| Response : Edit : Delete |
|
Hi, English documentation for the new OPEN-R SDK v1.1.5 has been released. You can download it via the Download page. Thank you, OPEN-R SDK Support |
| V 0999 OPEN-R SDK 1.1.5 released openrsupport@(Administrator) [2004/03/02-08:00] |
| Response : Edit : Delete |
|
Hi everyone, We are pleased to announce the immediate availability of OPEN-R SDK 1.1.5. This version supports both ERS-7 and ERS-200 series robots. Please note: * The Memory Stick images in the new SDK are different, for ERS-7 and ERS-200 series, because of optimization for each CPU. Please take care to use the correct Memory Stick image for your robot. * cygwin and gcc versions are updated. Please use the SDK with newer cygwin and MIPS cross development environments. * cygipc is now included in cygwin -- as a result, /user/local/bin/ipc-daemon.exe is renamed to /usr/bin/ipc-daemon2.exe. * English documentation and model information will be released soon. Thank you, OPEN-R SDK Support |
| V 0995 The robot shuts down when I run the OPEN-R Program!!! xiaozhu [2004/02/27-12:47] |
| | Response : HomePage : Email : Edit : Delete |
|
| My
AIBO Entertainment Robot ERS-7 works smartly when I communicate it with | touch, AIBO Cards, WLAN( to view the photos) . However, it shuts down when I | try to run the HelloWorld Program (which is one of the sample programs). | The original Memory Stick in my AIBO Entertainment Robot ERS-7 is ERA-MS032. | I copy all the necessary files to the Memory Stick and then boot the robot, but | the robot shuts down automatically. | What is the problem? Is it because that the ERA-MS032 Memory Stick can not be | used for OPEN-R Program? | |
| +- 0996 RE:The robot shuts down when I run the OPEN-R Program!!! AiboPet [2004/02/28-02:35] |
| Response : Edit : Delete |
|
>
What is the problem? Is it because that the ERA-MS032 Memory Stick can
not be >used for OPEN-R Program? Exactly! You need a "PMS" Programmable Memory Stick. Part number ERA-MS016 (with no ERF- sticker on it) See the FAQ: https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq5.php4#D0008 |
| V 0990 What folders and files should be copied? xiaozhu [2004/02/25-13:51] |
| | Response : HomePage : Email : Edit : Delete |
|
|
What folders and files should be copied to the AIBO Programming Memory | Stick if I want to run my own program on AIBO ERS-7? | How should they be arranged? | |
| +- 0991 RE:What folders and files should be copied? sapient [2004/02/27-00:45] |
| Response : HomePage : Email : Edit : Delete |
|
>
What folders and files should be copied to the AIBO Programming Memory >Stick if I want to run my own program on AIBO ERS? Use the command # cp r <directory of installation>/OPEN_R_SDK/OPEN_R/MS_ERS7/WCONSOLE/ memprot/OPENR <destination> The directory where you installed the sdk is probably usrlocal on linux or mac not sure for windows The destination is your memory stick This will have put aperios, and the system files on the memory stick now compile your program make, make install then copy the MW MiddleWare directory A command like this should dofor the sample programs the MS file is in the first directory of the programsame as the makefile # cp r MS/OPEN-R/MW <destination>/OPEN-R Thats all for the basic programs Put the memory stick in your aibo & turn it on oThe article that posted, was modified by Author (The article that 2004/02/27-00:44 posted, was modified by Author) |
| V 0992 Which type of Memory Stick can be used for OPEN-R? xiaozhu [2004/02/26-12:48] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
see lots of Memory Stick : MSA-32A, MSH-M128A, MSX-M256A, PCGA-MM512U,
ERA- | MS016 .... | Which type can be used for OPEN-R in AIBO ERS-7? Is other type compatible? | |
| +- 0994 RE:Which type of Memory Stick can be used for OPEN-R? l337Nick [2004/02/27-00:39] |
| Response : Edit : Delete |
|
ERA-MS016
is the one you want and you'll need something like MSAC-PC1 or MSAC- PC2 as a reader/writer |
| V 0989 AIBONET montgrimpulo [2004/02/25-05:38] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I'm using Apple's OS X 10.3.2 and my mac is connected via an ethernet cable to an Apple airport base | station V5.3. My Aibo (210) has a LAN card. | | I want to set up a wireless connection to my AIBO. What are the parameters for System Preferences: | Network ? The Airport is currently used to connect mine and other macs (wireless) to the Internet. Do I | have to change anything for the Airport using Aiport Admin Utility? What do I have to set in | WLANDFLT.TXT ? | How can I easily test if my AIBO is connected ? | Thanks | |
| +- 0993 RE:AIBONET sapient [2004/02/26-19:43] |
| Response : HomePage : Email : Edit : Delete |
|
>I'm
using Apple's OS X 10.3.2 and my mac is connected via an ethernet cable
to an >Apple airport base station V5.3. My Aibo (210) has a LAN card. > >I want to set up a wireless connection to my AIBO. What are the parameters for System >Preferences: Network ? The Airport is currently used to connect mine and other macs >(wireless) to the Internet. Do I have to change anything for the Airport using Aiport >Admin Utility? What do I have to set in WLANDFLT.TXT ? You don't need to set anything in WLANDFLT.TXT, but you must add the WLANCONF,TXT file SYSTEM/CONF... My copy looks a bit like this: HOSTNAME=aibo3 ETHER_IP=10.0.1.100 ETHER_NETMASK=255.255.255.0 IP_GATEWAY=10.0.1.1 ESSID=NAME OF AIRPORT WEPENABLE=1 WEPKEY= APMODE=1 CHANNEL=10 >How can I easily test if my AIBO is connected ? open a terminal and type: # ping <aibo IP> or # telnet <aibo IP> 59000 This is just out of my head and may not all be right. I suggest you look at a tutorial on the subject like this one by Eric Vaughan: http://www.droidlogic.com/aibo/open-r/OSX.html which takes you step-by-step through everything. A |
| V 0987 Author DELETE ***** [2004/02/24-09:54] |
| Response : Edit : Delete |
|
(The
article that 2004/02/23-17:43 posted, was modified by Author) |
| V 0979 Are there any documents on sample programs? zhangtao [2004/02/18-11:41] |
| | Response : Email : Edit : Delete |
|
| Hello all, | | I am just learning how to operate AIBO and develop some programs by OPEN_R | SDK. I have downloaded the sample programs. But I did not find any documents | on these sample programs. Could you tell me how I can get the documents on | sample programs from internet? | | Thank you. | |
| +- 0981 RE:Are there any documents on sample programs? TodboT [2004/02/19-03:19] |
| | Response : Edit : Delete |
|
| There
isn't really any good documentation in English for the sample programs. | If you can read Japanese, there's an OPEN-R book. | | There are the programmers and level 2 reference guides that come in the | documents package in the download section. | | And there is a pretty good OPEN-R tutorial here: | | http://www.ensta.fr/~baillie/openr_tutorial.html | | Tod | |
| +- 0988 RE:Are there any documents on sample programs? zhangtao [2004/02/23-17:55] |
| Response : Email : Edit : Delete |
|
>There
isn't really any good documentation in English for the sample programs. >If you can read Japanese, there's an OPEN-R book. > >There are the programmers and level 2 reference guides that come in the >documents package in the download section. > >And there is a pretty good OPEN-R tutorial here: > >http://www.ensta.fr/~baillie/openr_tutorial.html > >Tod Your suggestions are very helpful for me. Thank you very much. |
| V 0986 RP and Subject on remote side Nasnah [2004/02/22-00:45] |
| Response : Edit : Delete |
|
The
tutorial #7 explains how to set up an object as an observer on the
remote side. But it does not explain how to set up a subject on the remote side. Imagine that I have two objects Commander and Perception. The connect.cfg before I use remote processing looks like: -------------------------------------------------------------------- Commander.SendImage.Image.S Perception.ReceiveImage.Image.O Perception.SendResults.Results.S Commander.RecieveResults.Results.O -------------------------------------------------------------------- Now the processing of an image takes a long time and I want to do it on a fast computer. So I use RP and run the Perception object there. The connect.cfg on Aibo now looks: -------------------------------------------------------------------- Commander.SendImage.Image.S TCPGateway.ReceiveImage.Image.O TCPGateway.SendResults.Results.S Commander.ReceiveResults.Results.O -------------------------------------------------------------------- On the host(the fast computer): -------------------------------------------------------------------- TCPGateway.SendImage.Image.S Perception.ReceiveImage.Image.O Perception.SendResults.Results.S TCPGateway.ReceiveResults.Results.O -------------------------------------------------------------------- The question now is how should the ROBOTGW.CFG and HOSTGW.CFG look like? My suggestion would be like this: ROBOTGW.CFG: ------------------------------------------- TCPGateway.Proxy.AperiosMessage.P 59001 TCPGateway.ReceiveImage.Image.O 59002 TCPGateway.SendResults.Results.S 59003 ------------------------------------------- HOSTGW.CFG: ----------------------------------------------- !ROBOT_PROXY 59001 192.168.x.x TCPGateway.SendImage.Image.S 59002 192.168.x.x TCPGateway.ReceiveResults.O 59003 192.168.x.x ----------------------------------------------- But this does not work. How should it look like? |
| V 0985 About abort() and overriding it curufin [2004/02/22-00:04] |
| Response : Edit : Delete |
|
I'm
having problems with the abort function When running my design om Aibo it executes properly for a while a couple of minutes and then the telnet console shows "abort called" and it seems one of the running objects dies Earlier I got the advice to override the abort function, but I'm not entirely sure how to do this not being familiar with this function I define a function like this void [classname]abort and include a SYSPRINT in it However, I get the same "abort called" message in telnet and my SYSPRINT doesn't appear It seems my overriding doesn't work Any advice on how to properly override the function would be appreciated Also, I'd very much like to know what may be causing the function to be called in the first place. (The article that 2004/02/22-00:03 posted, was modified by Author) |
| V 0984 RP-OPENR compiling problems vf2001 [2004/02/21-04:20] |
| Response : Email : Edit : Delete |
|
I have installed the newest version of OPEN_R_SDK. When running the RP setup-rp-config I get a couple of errors in the TCPGateway object and the OService object due to shared memory. Any idea? thanks! |
| V 0983 abort() called kinhouse [2004/02/20-03:47] |
| Response : Edit : Delete |
|
I
got abort() called message than my program down. However, the dog still moving but can't sent or receive any message. I would like to know what is abort() called, and what cause this? My program run smoothly at the beginning, but after some time, may be 20 s, less or more, abort() called displayed. I can't solve this problem as I ahve no idea where is come from. Anyone can help? Thx a lot |
| V 0976 MoNet sapient [2004/02/17-01:01] |
| | Response : HomePage : Email : Edit : Delete |
|
| Hello all, | | I'm trying to use MoNet for my own projects on an ERS7. It seems like it's possible to | modify/add your own functions to MoNet by adding more MTN files to the ODA file. Is this | correct? If so are there any resources on this. I've looked on the internet but haven't | found anything that says anything about doing this. | | I am already sending MoNet walk commands but they seem to interfere with my home | made head control algrithm. Is there any easy way to control the walk with MoNet, while | still controlling the head by myself. | | Thanks for your interest. | | A | | |
| +- 0977 RE:MoNet AiboPet [2004/02/17-02:23] |
| | Response : Edit : Delete |
|
| >
I'm trying to use MoNet for my own projects on an ERS7. It seems like
it's | possible to | >modify/add your own functions to MoNet by adding more MTN files to the ODA | file. Is this | >correct? | | Correct. | You need to add the MTN files to the MOTION.ODA file and also tweek the | MONETCMD.CFG file to include the new motion names (assuming you use that | technique). I don't think you need to touch the MONET.CFG | | The motion file names should be in the proper format (eg: | a_startposture#endposture_blahblahblah.mtn) | | --- | The Skitter motion editor can be used to create .MTN files (recommended). | Or you can steal them from the commercial AiboMind MOTION.ODA (lots in there) | | I have a number of DOS command line tools for mucking around with these files: | http://www.aibohack.com/a/cmd_line_tools.zip | Extracting MTN files from ODAs is done with DUMPODA | Building ODA files from MTN files is done with MAKEODA | | ----- | > I am already sending MoNet walk commands but they seem to interfere with my | home | >made head control algrithm. Is there any easy way to control the walk with | MoNet, while | >still controlling the head by myself. | | The sample walk motions are full-body motions (a_walk#walk_blah.mtn) | The 'a_' means the motion uses all the servos. | | One solution would be to extract the 12 leg servos and save it as an 'l_' | motion (eg: "l_walk#walk_blah2.mtn"). I believe that is supported by the MoNet | sample ('l_' = legs only) | |
| +- 0978 RE:MoNet sapient [2004/02/17-17:50] |
| Response : HomePage : Email : Edit : Delete |
|
Thanks ever so much. That was the confirmation I needed. I think that all the motions that I'll need will be in the Aibo mind software. I found your tools while searching around on the net, but I only have access to Macs & linux. Has anyone ported the scripts to bash or applescript (& would they be appriciated?) |
| V 0935 ERS7 camera focal length? hcklbrryhnd [2004/01/28-01:26] |
| | Response : Email : Edit : Delete |
|
| Hi, | Are the camera parameters for ERS7 the same as that for ERS210A? We know that | the opening angles have changed slightly and the image dimensions are larger. | But what about things like the focal length, gain, white balance etc? | Also, for the ERS7, does anybody know the pixels/mm resolution that we get?? | | Mohan.S. | |
| +- 0939 RE:ERS7 camera focal length? openrsupport@(Administrator) [2004/01/29-02:41] |
| | Response : Edit : Delete |
|
| Hello, | | An updated version of the ERS-7 hardware specifications will be posted | soon. But to answer your questions now, here are some specs for the | ERS7 camera. | | CMOS part: 1/4 inch | Number of picture elements: 416(H) x 320(V), 30FPS | Lens: F2.8, f = 3.27mm | Angle of view: | Horizontal angle: 56.9 deg | Vertical angle: 45.2 deg | Default: | White balance: 5000K fixed | Shutter speed: 1/100 sec fixed | Gain: 0dB fixed | | Thank you, | | OPEN-R SDK Support | |
| +- 0971 RE:ERS7 camera focal length? kcph007 [2004/02/14-12:39] |
| | Response : Edit : Delete |
|
| >Number
of picture elements: 416(H) x 320(V), 30FPS | | Hi, | How can it get to 416x320, when i set to HIGH RESOLUTION, it's only 208x160, did | I miss anything? | Thank you. | Kim | |
| +- 0973 RE:ERS7 camera focal length? ejt [2004/02/15-07:53] |
| | Response : HomePage : Email : Edit : Delete |
|
| Take
a look at the W3AIBO sample code - there's two versions of the
ReconstructAndConvertYCbCr | function at the end of JPEGEncoder.cc that will take the y-channel and the 3 y-channel derivates | and form a single, double resolution y-channel image. | | Note that this only works for the Y-channel... for the chromaticity channels (U and V, aka Cb and | Cr?) there's no extra information to do recover the original resolution. | | I've also adapted this code in the Tekkotsu framework btw... | http://www-2.cs.cmu.edu/~tekkotsu/dox/RawCameraGenerator_8cc-source.html#l00377 | which you can access it using the highest layer index of the FilterBankEvents... | I should point there may be a buffer overrun going on in RawCameraGenerator... (as | referenced in a previous thread) that's still to be looked into more. | | -ethan | |
| +- 0974 RE:ERS7 camera focal length? kcph007 [2004/02/15-18:08] |
| | Response : Edit : Delete |
|
| Thank
you for answering, ethan. | I've noticed that sample before (in BMP.cc as well). But what's the point of | contructing a bigger grey-scale image? Apart from displaying it, is there any | other purpose do you know? | | Kim | |
| +- 0975 RE:ERS7 camera focal length? AiboPet [2004/02/16-05:18] |
| Response : Edit : Delete |
|
>...But
what's the point of >contructing a bigger grey-scale image? Apart from displaying it... For computer calculations (on the AIBO itself), you can use the bigger grey scale image to give you greater resolution for object detection. I *believe* the Evolution Robotics object detection logic of the ERS-7 AiboMind software uses the higher res data. This lets them pick out object features at a much greater distance. For displaying to a human, you can construct a high-res *color* image using the color information as well (four pixels share the same U and V settings) The VisionU module of AiboSpeedRacer does this and JPG compresses the results: http://www.aibohack.com/openr_sdk/speedracer.htm (see the 416x320 color image at the top of the page) |
| V 0960 Problems with large (~100KB+) region allocation? ejt [2004/02/06-15:14] |
| | Response : HomePage : Email : Edit : Delete |
|
| We've
been running into this now and then for a while... Tom and Alok (others
in | our group) have run into odd problems in the past when working with large | regions of data, and recently I have as well. It's hard to pin this down as a | system library problem, but in all of our cases, switching from a dynamically | allocated region to a statically allocated region solved the problem. | | Since the malloc libraries appear to be supplied by libapsys, which I believe is | an Aperios library, I'm afraid I must hand my search for the bug over to you. | In order to help you track this down, I've spent a fair bit of time trying | various experiments to find a pattern. | | I ran into the problem when computing double-resolution images. (reconstructed | Y-channel from the Y-derivative channels, or simply upsampled versions of the | other channels for consistency) As you know, on an ERS-210, one of these images | is 352*288, or 99KB. On an ERS-7, its 416*320=130KB. I have done most of my | testing on the ERS-210, but the same problems occur on an ERS-7. | | What happens is that, commonly, the Aibo will crash when trying to free the | region(s). This can either happen when shutting down, or when completing saving | a file to memory stick from the CameraBehavior. | I have included a crashDebug run here: | http://tekkotsu.no-ip.org/media/memcrashdebug.txt | I haven't had a crash while saving to memstick recently, but the crashes on | shutdown are almost (but not *quite* 100%) consistent. There may be some | dependance on the layout of memory and order of allocations in order to trigger | the crash. | | However, in the Tekkotsu 2.0.1 release, I switched to using a staticly defined | memory region to hold the double resolution images, and everything then began to | work just fine. | | Since 2.0.1 includes several other bug fixes, I have generated a (very simple) | patch to switch RawCameraGenerator back to using dynamic allocation. This way | you can run tests with the minimal necessary changes to trigger the bug. | http://tekkotsu.no-ip.org/code/RawCameraGenerator.diff | For reference, the command to apply in Linux would be: | cd Tekkotsu/Vision; patch -p0 < RawCameraGenerator.diff | or you could just read the file and do it by hand since it's only a few lines | different. | | I have a couple log dumps of sample runs to illustrate the problem. Although | there isn't anything conclusive, each 'double size layer...' represents an | attempted allocation, yet only the first causes an actual decrease in free | memory (as reported by AnalyzerGetSizeOfFreeMemory() - this is accessed through | Free Memory Report, under the Status Reports menu) | The first allocation causes a decrease of rougly 400KB of free memory, even | though only 100KB is being requested. Even if this is simply a bug in the | reporting of free memory, the real problem is that the Aibo will crash when | those region(s?) are free()'d. | http://tekkotsu.no-ip.org/media/membugrep.txt | http://tekkotsu.no-ip.org/media/membugrep2.txt | I was triggering individual allocations by using the !set commands in the | Controller to switch the rawcam_encoding configuration between different | channels. In membugrep.txt I step through all 6 channels, in membugrep2.txt I | switch from y-only to full color (y, u, and v channels) | | I would also like to point out that a newer version of newlib (which has malloc | functions available) has been released. I tried updating to it with minor | success, but when I realized that malloc libraries were coming from libapsys I | gave up since it wouldn't fix the problem at hand. Perhaps someone over there | can look into using the newer libraries as a solution? | | Thanks! | -ethan | |
| |- 0961 RE:Problems with large (~100KB+) region allocation? sapient [2004/02/09-19:14] |
| | Response : HomePage : Email : Edit : Delete |
|
| Thanks for your post, | | Although I haven't any ideas on solving this problem, it has occoured in my code aswell | and your post saved me from wasting lots of time working out what was wrong... | | :-) | | A | |
| +- 0965 RE:Problems with large (~100KB+) region allocation? openrsupport@(Administrator) [2004/02/11-09:45] |
| | Response : Edit : Delete |
|
| Hello, | | Thank you for your very detailed bug report!! We analyzed your post | and determined the following. | | 1) We did not find any problem with the malloc(), fill() and free() | calls. | | 2) The exception in question occurred in the free() call. | | 3) The memory allocation header table was invalid(broken) | when the free() function tried to free the memory area. | | 4) We suppose that your program wrote over some data | in the memory allocation header table area. An overrun | write could cause this problem. | | For example : | | A = malloc(8192); | B = malloc(8192); | memset(A, 0, 8192 * 2); | free(A); | free(B); // exception | | These types of overrun bugs are typically challenging to fix. | | Thanks, | | OPEN-R SDK Support | |
| +- 0966 RE:Problems with large (~100KB+) region allocation? ejt [2004/02/11-11:19] |
| | Response : HomePage : Email : Edit : Delete |
|
| Thanks
for checking that out, I'll try to look into it some more on my end...
I'd studyied that section | of code for a while and everything looked OK, but it's not impossible there's a bug there. I'll pad | the region with some magic values and see if they get overwritten. | | In the mean time, could you perhaps give some more detail on the function I was using to get the | free memory? (AnalyzerGetSizeOfFreeMemory) I think Alok came across that a while back and it | appears undocumented. But it would be nice if we understood it better so we could watch for | memory leaks or plan for how much memory is available for certain tasks... however, currently the | results it gives are a little odd. | | thanks! | -ethan | |
| +- 0969 RE:Problems with large (~100KB+) region allocation? openrsupport@(Administrator) [2004/02/13-06:55] |
| | Response : Edit : Delete |
|
| Hi, | | "AnalyzerGetSizeOfFreeMemory" is not part of the OPEN-R API. | Typical OPEN-R applications should not use this function. | The "AnalyzerGetSizeOfFreeMemory" function bypasses the heap | library (free, malloc) and directly accesses the OS memory manager. | | The AnalyzerGetSizeOfFreeMemory() function reports the total memory | size that the heap library can get from the OS memory manager. | | But the free memory(in system) is fragmented, so malloc() may fail | even if AnalyzerGetSizeOfFreeMemory() says that the free size is larger | than the amount that you want to malloc. | | When you call malloc() for the first time, the heap library gets a | memory area from the operating system. Typically, this size is larger | than the size of malloc() and only a part of the memory area is given | to the application program layer. | | When you call malloc() the 2nd time, malloc may give the memory from the | area that is managed by the heap library. (system memory area does not | decrease). | | Even when you call free(), the heap library might hold onto the memory area. | This means that the system's free memory does not increase. | | As a result: | AnalyzerGetSizeOfFreeMemory() is useful only for the system layer. | Looking from the application program layer, AnalyzerGetSizeOfFreeMemory() | may return odd values. | | Thanks, | | OPEN-R SDK Support | |
| +- 0972 RE:Problems with large (~100KB+) region allocation? ejt [2004/02/15-02:56] |
| Response : HomePage : Email : Edit : Delete |
|
Thanks!
That makes much more sense now. :) |
| V 0967 Critical Section Support phalanx [2004/02/12-14:51] |
| | Response : Email : Edit : Delete |
|
| I
was just wondering if there exists support for critical sections in
OPEN-R. | Often you have different sections of code driven in the same memory space by | the signals from, for example, incoming vision frames and incoming ready | signals for joint movement. | Critical section support (and support for threads in general) would greatly | improve OPEN-R. You dont always want to use the entire overhead of the OPEN-R | object framework every time you want a new thread. | | Thanks for your consideration. | | Nathan | |
| +- 0968 RE:Critical Section Support ejt [2004/02/13-02:55] |
| Response : HomePage : Email : Edit : Delete |
|
Ooo,
ooo! I'll chime in my support for this one... We had to write our own mutual exclusion lock, but I'm *sure* this could be done much more efficiently at the OS level, and looking through the OPENR headers I've seen some references to getting the status to locks, but I don't know how to use them. Anyway, in the mean time, interested parties can use this: http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/Shared/MutexLock.h?rev=1.9&content-type=text/vnd.viewcvs-markup It's pretty well tested. Don't take my word for it though - i'd want to know if you get any odd results from it. I think you can take out the header files which are included btw. (debuget.h just has some ASSERT macros, which it doesn't look like I have any in there any more...) |
| V 0937 Walk a certain distance shivangp [2004/01/28-08:46] |
| | Response : Email : Edit : Delete |
|
| I
wanted to know how I can modify the MotionCommander/SoccerLion sample
to walk | a certain distance? | |
| +- 0964 RE:Walk a certain distance openrsupport@(Administrator) [2004/02/11-09:38] |
| Response : Edit : Delete |
|
Hi, SoccerLion is a sample to show how to use many different OPEN-R objects to connect with each other. SoccerLion uses commercial OPEN-R objects, so the source code is not open. That means the sample is difficult to modify. You can learn how to connect objects and how to have objects cooperate from the SoccerLion sample, but it is not a good sample to be a base application. The MoNet sample is a good sample program that demonstrates walking a certain distance. If you need a more complicated program, the Tekkotsu framework or the source code from one of the RoboCup teams would be a good example. Thanks and good luck, OPEN-R SDK Support >I wanted to know how I can modify the MotionCommander/SoccerLion sample to walk >a certain distance? |
| V 0950 Broken ERS-7 ? Sauber [2004/02/03-15:29] |
| | Response : Email : Edit : Delete |
|
| Hi, | We appear to have a broken ERS-7. The robot was working fine. All we did was | turn the robot off and change the memory stick. Now the robot no longer boots at | anytime - with or witout a memory stick. | | We have tested the robot with multiple batteries and sticks. | | Our code generates the following log file - | | context: 80234740 | object: ovirtualRobotComm | badvaddr: 00000000 | epc: 80468568 | ra: 80468e1c | target address: 80468568 (epc) | gp: 803b46a0 | _gp: 00528340 | static addr: 005dc208 | symbol: 00528340 A _gp | | We assume it's time for warranty service :). | | Michael & Craig | NUbots (University of Newcastle) | |
| |- 0952 RE:Broken ERS-7 ? GTleader [2004/02/03-20:51] |
| | Response : HomePage : Email : Edit : Delete |
|
| Although
all our ERS-7 still boot, all of them have at least one leg joint | that make a strange noise when we move it manually while the robot is switched | off. One of the robots even has a joint that makes such a noise when it walks. | There does not seem to be a pattern, which joints show this effect. | | Are there any other teams having the same problem? | | Thomas (GermanTeam) | |
| +- 0963 RE:Broken ERS-7 ? openrsupport@(Administrator) [2004/02/11-09:34] |
| Response : Edit : Delete |
|
Hello, We are sorry to hear about the troubles you are having. Our suspicion is that it is indeed a hardware problem, but we cannot definitively diagnose your problem via this forum. We recommend that you send the AIBO in for service. Thank you, OPEN-R SDK Support >Hi, > We appear to have a broken ERS-7. The robot was working fine. All we did was >turn the robot off and change the memory stick. Now the robot no longer boots at >anytime - with or witout a memory stick. > >We have tested the robot with multiple batteries and sticks. > >Our code generates the following log file - > >context: 80234740 >object: ovirtualRobotComm >badvaddr: 00000000 >epc: 80468568 >ra: 80468e1c >target address: 80468568 (epc) >gp: 803b46a0 >_gp: 00528340 >static addr: 005dc208 >symbol: 00528340 A _gp > >We assume it's time for warranty service :). > >Michael & Craig >NUbots (University of Newcastle) |
| V 0946 RCode for ERS7? kevinkw [2004/02/01-13:17] |
| | Response : Edit : Delete |
|
| Anyone have any idea when R-Code for the ERS7 will come out? | I'm a middle school CS teacher and the Sony Rep at the Meteron said that R-Code | was for all Aibos. Seems not to be so once I visited the Sony US Aibo site. | Also are there any CS teachers out there who have used the Aibo as a teahcing | tool? | |
| |- 0953 RE:RCode for ERS7? slider2nl [2004/02/03-23:30] |
| | Response : Edit : Delete |
|
| >Anyone
have any idea when R-Code for the ERS7 will come out? | >I'm a middle school CS teacher and the Sony Rep at the Meteron said that R-Code | >was for all Aibos. Seems not to be so once I visited the Sony US Aibo site. | >Also are there any CS teachers out there who have used the Aibo as a teahcing | >tool? | | I read somewhere that it was begin februari.. | |
| +- 0962 RE:RCode for ERS7? openrsupport@(Administrator) [2004/02/11-09:32] |
| Response : Edit : Delete |
|
Thank
you for your interest. At this time, we have not yet announced the availability of R-CODE for ERS-7. Should this change, we will post about it here. Thank you, OPEN-R SDK Support >Anyone have any idea when R-Code for the ERS7 will come out? >I'm a middle school CS teacher and the Sony Rep at the Meteron said that R-Code >was for all Aibos. Seems not to be so once I visited the Sony US Aibo site. >Also are there any CS teachers out there who have used the Aibo as a teahcing >tool? |
| V 0921 Author DELETE ***** [2004/02/09-03:18] |
| Response : Edit : Delete |
|
(The
article that 2004/01/25-00:41 posted, was modified by Author) |
| V 0877 JAM GTleader [2004/01/09-04:11] |
| | Response : HomePage : Email : Edit : Delete |
|
| While
developing new kicks for the ERS-7, we very often get the message | | [oid:80000038,prio:1] OVirtualRobotComm : JAM. SHUTDOWN... | | A few questions: | - Why does that happen? | - How to avoid it? | - Is there any chance to avoid that the robot is switched off? This is a quite | extreme reaction! | | Thomas | |
| +- 0878 RE:JAM ejt [2004/01/09-07:10] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
would tend to agree the threshold could be raised a little. Now that
i've | worked with the ERS-7 some more, I've gotten the jam condition shutdown a couple | times when there really wasn't *that* much pressure on the joints. It could | probably stand allowing just a little more pressure I think. | | After all, I'm even getting this because of a little twitch at startup, even | though the joints are free to move... (see previous post) | |
| |- 0879 RE:JAM gijoe [2004/01/09-11:34] |
| | Response : HomePage : Email : Edit : Delete |
|
| We
(NUbots) have experienced the same problem. | |
| +- 0880 RE:JAM GTleader [2004/01/09-17:20] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
think the best solution would be that the program can "measure" the JAM- | condition in some way and the robot just sets the pid-values of the relevant | joints to zero to protect itself. In such a case, the program can detect the | situation and handle it appropriately, e.g. by learning to avoid jamming | combinations of joint angles. | | Thomas | |
| +- 0954 RE:JAM GTleader [2004/02/04-02:37] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
think, I just repost this message, because an answer is still missing.
Is | there anything Sony will do about this situation? | | Thomas (GermanTeam) | |
| +- 0959 RE:JAM openrsupport@(Administrator) [2004/02/05-09:45] |
| Response : Edit : Delete |
|
Hello, We are monitoring these posts, and we are considering this jam detection threshold issue. Unfortunately, we do not have an answer at this moment. Should this change, we will inform you as soon as possible via this BBS. Thank you, OPEN-R SDK Support |
| V 0956 2x0 vs 2x0A Processing Power Marizu [2004/02/05-04:39] |
| | Response : Edit : Delete |
|
| I
was wondering how much extra processing power (approx %) a
supercore(2x0A) | gives over a 2x0 series core. | I'd like to run a multilayer feedforward neural network in one (not the | training). | | Cheers, | Mari-zu | |
| +- 0957 RE:2x0 vs 2x0A Processing Power ejt [2004/02/05-07:00] |
| | Response : HomePage : Email : Edit : Delete |
|
| It's
a pretty significant increase - the supercore really is about twice as
fast. | To compare in particular the time required to process one camera frame in | Tekkotsu v2.0: | Exponential Overall | Average Average | ERS-210: 20.6 ms 20.0 ms | ERS-210A: 10.5 ms 10.0 ms | ERS-7: 3.4 ms 3.1 ms | | Vision processing is mostly integer math, compared to the time required to send | a buffer of 4 motion frames to the system (which has more floating point): | Exponential Overall | Average Average | ERS-210: 2.6 ms 2.5 ms | ERS-210A: 1.4 ms 1.5 ms | ERS-7: 0.7 ms 0.8 ms | | Times are wall-clock, not CPU time, so they can be thrown off by multitasking, | but are sampled over 5 minutes of run time, so it's a pretty large sample size | (~7000 frames on a ERS210 (runs @25 FPS) and ~8000 frames on a ERS7 (runs @30FPS)) | | What I find of interest in these results is that even though the clock speed on | the ERS-7 is only 50% faster than the 210A (576MHz vs 384MHz), it's turning in | results that are consistently twice as fast. But don't forget the ERS-7 is also | running a beta of a newer OS, so that might also signal a more general speedup | when Sony releases the next version for all models. | | You can check see the full benchmark results across models (including the ERS-7) | from several profiled sections of our code, available at: | http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/docs/benchmarks/ | | In case you're curious about the profiling code that generates these reports, | which is quite portable, so may be useful even if you aren't using the rest of | the framework, there's documentation here: | http://www-2.cs.cmu.edu/~tekkotsu/dox/classProfiler.html | and source here: | http://www.tekkotsu.org/dox/Profiler_8h-source.html | http://www.tekkotsu.org/dox/Profiler_8cc-source.html | |
| +- 0958 RE:2x0 vs 2x0A Processing Power ejt [2004/02/05-07:06] |
| Response : HomePage : Email : Edit : Delete |
|
I
should add that if you want to save the Profiler source, use the CVS
interface: http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/Shared/Profiler.h?rev=1.10&content-type=text/vnd.viewcvs-markup http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/Shared/Profiler.cc?rev=1.12&content-type=text/vnd.viewcvs-markup otherwise you'd have to copy and paste and then delete all the line numbers (eww) Or of course, you can always just download the whole framework at once... http://www.tekkotsu.org/TekkotsuInstall.html |
| V 0948 YUV oddity? (U and V swapped, maybe?) ejt [2004/02/03-05:34] |
| | Response : HomePage : Email : Edit : Delete |
|
| Maybe
I'm just going crazy, but it seems sometimes U and V are swapped from
what | they should be... but sometimes not - it's hard to tell since i'm using code i | didn't write, so I don't completely have my head around it. | | Figured i'd check if anyone else had noticed this before I go trying to hunt | this down or swap which channel is which in our own code... | | thanks, | -ethan | (The article that 2004/02/03-05:27 posted, was modified by Author) | |
| +- 0949 nevermind... ejt [2004/02/03-06:20] |
| | Response : HomePage : Email : Edit : Delete |
|
| No
bug, just an odd interface with libjpeg that made me doubt what I was
doing | was right. | | Just for reference, in case anyone else ever gets confused - the jpeg libraries | take data in the form YCbCr, instead of the YCrCb that I would assume (since | that corresponds to YUV). | On the decompresson side, if you extract individual components from a JPEG image | in java, the same thing happens - component 0 is Y, component 1 is Cb, component | 2 is Cr. | | If anyone knows why they can't pick an ordering convention and stick with it, | email me ;) | | *shrug* | |
| +- 0955 RE:nevermind... AiboPet [2004/02/04-03:14] |
| Response : Edit : Delete |
|
FWIW:
I am still interpreting the data as YUV (YVU actually) instead of YCrCb (YCbCr) YUV and YCrCb are similar but have slightly different conversion to RGB. The documentation says the camera data is "YCrCb", but I found the colors looked better sticking with YUV encoding/transformation. I tried the YCrCb/YCbCr encoding, but the colors didn't look as good compared to the actual object under AIBO's nose. |
| V 0899 Soccer Lion source code deadmanhead [2004/01/16-00:53] |
| | Response : Email : Edit : Delete |
|
| I
have a project for RoboCup recently. i found Soccer Lion is useful
reference. | But some codeing e.g OMCDT.bin that i can't find the sources. and i find a file | (libOMWares.a )in lib folder. Is it a source code? if not, how can i find it ? | |
| +- 0914 RE:Soccer Lion source code openrsupport@(Administrator) [2004/01/21-13:46] |
| | Response : Edit : Delete |
|
| I'm
sorry, but most of the source code for SoccerLion is not available. | | SoccerLion is a sample program that shows how to connect many objects together | for application purposes. | | SoccerLion uses commercial based middle ware objects. | | Please understand that some parts of the technology cannot be disclosed because | of commercial business reasons. | | Thank you, | | OPEN-R SDK Support | |
| +- 0943 RE:Soccer Lion source code deadmanhead [2004/01/31-16:38] |
| | Response : Edit : Delete |
|
| I
understand your point. but recently i have a academic project about
robocup. | i relly want some help for understand SoccerLion. And my duty is focus on | vision. So, i want to know more about how does the OMCDT work. Can i find some | reference about this? For example, what are parameters in "CdtStruct" talking | about? Thanks for you help! | |
| +- 0944 RE:Soccer Lion source code AiboPet [2004/02/01-03:01] |
| | Response : Edit : Delete |
|
| >i
relly want some help for understand SoccerLion. And my duty is focus on | >vision. So, i want to know more about how does the OMCDT work. | | General advice: | My suggestion is to look at the header in question (CdtStruct.h). Although not | documented explicitly, you can usually figure things out by looking at the | header, and the sample code (and assuming you are generally familiar with how | OpenR objects communicate) | This will usually help you understand what is currently being done, but may | not help if you want to fully exploit the undocumented module. | | --- | In the case of the color detection, the functionality is very simple | Setting the color detection ranges is similar to the documented OpenR SDK way | of doing things. | Getting the color detection result (ie. where it thinks the pink ball is) is | relatively simple, using their own object types. | | If you want to do more elaborate processing, my suggestion is to look at the | BallTrackingHead example. It does similar tracking in the sample itself (not | in another module) | IMHO "OMCDT" is not the place to start for a RoboCup quality vision module | |
| +- 0947 A bit more advice... sapient [2004/02/03-01:23] |
| | Response : HomePage : Email : Edit : Delete |
|
| Most
robocup teams publish their code. | Try thier websites, a good standard might be Tekkotsu, which is maintained by Carnegie | Mellon. | | http://www-2.cs.cmu.edu/~tekkotsu/index.html | | Also I know that RunSWift have won it many years over and publish all thier code. | | http://www.cse.unsw.edu.au/~robocup/ | | I think that in robocup you may have to accept that "imitation is the sincerely form of | flattery". Check the rules for authorship and make sure that you are *absolutely* clear | about who wrote what... | | Should be a good starting point. | | Hope that this helps. Soccerlion is difficult to hack. | | A | |
| +- 0951 RE:A bit more advice... deadmanhead [2004/02/03-20:32] |
| Response : Email : Edit : Delete |
|
thanks
for helping i think those staffs are very useful (The article that 2004/02/03-20:31 posted, was modified by Author) |
| V 0941 Updated sample programs for ERS-7 openrsupport@(Administrator) [2004/01/30-09:57] |
| | Response : Edit : Delete |
|
| Hello, | | The sample programs for OPEN-R SDK for the ERS-7 (beta2) | have been released. You can find them on the download page. | | Thank you, | | OPEN-R SDK Support | |
| +- 0945 RE:Updated sample programs for ERS-7 AiboPet [2004/02/01-03:04] |
| Response : Edit : Delete |
|
>The
sample programs for OPEN-R SDK for the ERS-7 (beta2) >have been released.... Friendly reminder: An update to the SoccerLion sample, and the necessary changes to OMWares middleware modules would be much appreciated. |
| V 0926 W3AIBO sample and jpeg-6b on Mac OS (Panther)? bob_ebert [2004/01/26-13:01] |
| | Response : Email : Edit : Delete |
|
| I'm
running into problems building the W3AIBO sample on Panther (Mac OS
10.3.2) | I'm having to mess around a bit to get the jpeg library to build. Well, it builds fine for MacOS, but I | think the resulting libjpeg.a will not be useful on MIPS. I'm guessing I probably need to build the | jpeg library using the mipsel-linux-gcc cross compiler instead, but I'm having trouble configuring | my system to do that. Am I on the right track? Any ideas? | | Short of that, anyone willing to share a compiled libjpeg.a binary? | |
| +- 0940 RE:W3AIBO sample and jpeg-6b on Mac OS (Panther)? openrsupport@(Administrator) [2004/01/29-12:23] |
| | Response : Edit : Delete |
|
| Please
read the README file in the W3AIBO sample folder. | | You should be able to put jpegsrc.v6b.tar.gz in the W3AIBO folder and type "make | install". | | The makefile does everything that you want it to do. | | OPEN-R SDK Support | |
| +- 0942 RE:W3AIBO sample and jpeg-6b on Mac OS (Panther)? bob_ebert [2004/01/30-14:10] |
| Response : Email : Edit : Delete |
|
>You
should be able to put jpegsrc.v6b.tar.gz in the W3AIBO folder and type
"make >install". Yes, that works now with the beta2 samples, thanks! I had also gotten it to work by building libjpeg.a on a PC and copying it over, but building from source feels better. |
| V 0928 Author DELETE ***** [2004/01/28-20:07] |
| Response : Edit : Delete |
|
(The
article that 2004/01/26-18:58 posted, was modified by Author) |
| V 0881 ERS-7 camera GTleader [2004/01/10-01:51] |
| | Response : HomePage : Email : Edit : Delete |
|
| The
camera of the ERS-7 is of low quality. Everyting close to the image
border | is blue. The camera of the ERS-210 was a lot better: | | http://www.informatik.uni-bremen.de/~roefer/ers7/ | | It seems that both of our robots have this problem. Do other owners/RoboCup | teams have the same problem? | | Thomas | |
| |- 0882 RE:ERS-7 camera ejt [2004/01/10-03:57] |
| | Response : HomePage : Email : Edit : Delete |
|
| >It
seems that both of our robots have this problem. Do other owners/RoboCup | >teams have the same problem? | Yup. We see it as well. I've been meaning to post about it as well, but | haven't gotten around to taking representitve sample images. | | This is a real pain for color segmentation, because colors at the edges are | distorted, so training the color thresholds becomes very problematic. I thought | the higher resolution of the ERS-7 would be good, but I'm afraid the color | issues will make it much harder to use. | | I suppose since the distortion is consistant, we could correct it in software, | but that'll require CPU power... :( | | -ethan | |
| +- 0934 RE:ERS-7 camera hcklbrryhnd [2004/01/28-00:58] |
| | Response : Email : Edit : Delete |
|
| >The
camera of the ERS-7 is of low quality. Everyting close to the image
border | >is blue. The camera of the ERS-210 was a lot better: | > | >http://www.informatik.uni-bremen.de/~roefer/ers7/ | > | >It seems that both of our robots have this problem. Do other owners/RoboCup | >teams have the same problem? | > | >Thomas | | Yes, I am seeing the same problem in our (UTAustin) robots (ERS7) too. Though | the image dimensions are larger, the corners of the image as you mention do | appear to have a blue shade. | |
| +- 0938 RE:ERS-7 camera bob_ebert [2004/01/28-11:45] |
| Response : Email : Edit : Delete |
|
>>It
seems that both of our robots have this problem. Do other owners/RoboCup >>teams have the same problem? Mine also has this problem. Looking closely, I can see the camera lens is recessed and has a shape that might introduce this distortion due to vignetting (it appears to be a fairly wide angle lens.) I wonder if the color is due to internal reflection or the curve of the plastic shell covering the head? (Anyone willing to try cutting/removing this cover and seeing if it makes a difference?) |
| V 0930 TCP Gateway for ERS-7 omega697 [2004/01/27-04:27] |
| | Response : Email : Edit : Delete |
|
| I've
been trying to port communication code from the ERS-210A to the ERS-7
and | every time I try to interface with TCP Gateway, the AIBO crashes immediately. | Is there any change in the TCP Gateway binary for the ERS-7? If so, where can I | get a new binary? | | Thanks, | | Kurt | |
| +- 0936 RE:TCP Gateway for ERS-7 openrsupport@(Administrator) [2004/01/28-03:50] |
| Response : Edit : Delete |
|
Hello, TCP Gateway and x86 binaries are included in the OPEN-R SDK main file for ERS- 7. Please see the download page. + OPEN-R SDK for ERS-7(only) Beta (6MB) OPEN_R_SDK-1.1.5-beta1.tar.gz (2003/11/04) Remote Processing OPEN-R is now included in the main OPEN-SDK distribution. Remote Processing OPEN-R includes executable binaries for Windows and Linux. These binaries are made with the following versions of gcc: Windows (Cygwin) gcc 3.2 Linux gcc 2.95 / 3.2 Important: Please follow the installation procedure outlined in the README in the archive file. And please consult the FAQ if you have any problem: https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq5.php4#D0007 Thank you, OPEN-R SDK Support |
| V 0885 Mac OS X (10.3.2) Panther Support vash [2004/01/12-06:40] |
| | Response : Email : Edit : Delete |
|
| I've
just joined and was wondering if anyone that has set up their
development | environment under Mac OS X (10.3 or earlier) would be kind enough to share any hints, | tips, traps, etc.? | | Anyone communicating with their aibo using an Airport terminal? Tips? | | Thanks! | |
| |- 0886 RE:Mac OS X (10.3.2) Panther Support ejt [2004/01/12-09:49] |
| | | Response : HomePage : Email : Edit : Delete |
|
| | >I've
just joined and was wondering if anyone that has set up their
development | | >environment under Mac OS X (10.3 or earlier) would be kind enough to share any | | yup. It's the same as a regular linux install - kudos to Sony giving us an OS X | | setup script... (build-devtools) :) | | | | I haven't used it with an Airport, but I have done peer-to-peer during demos by | | enabling my iBook to serve as a base station. I think I left the infrastructure | | mode setting in the Aibo's wlandflt.cfg even though I think in that case it's | | actually technically in p2p/ad-hoc mode... *shrug* it just worked, i didn't have | | to toy with it. | | |
| | +- 0887 RE:Mac OS X (10.3.2) Panther Support vash [2004/01/13-11:10] |
| | Response : Email : Edit : Delete |
|
| >>I've
just joined and was wondering if anyone that has set up their
development | >>environment under Mac OS X (10.3 or earlier) would be kind enough to share any | >yup. It's the same as a regular linux install - kudos to Sony giving us an OS X | >setup script... (build-devtools) :) | > | >I haven't used it with an Airport, but I have done peer-to-peer during demos by | >enabling my iBook to serve as a base station. I think I left the infrastructure | >mode setting in the Aibo's wlandflt.cfg even though I think in that case it's | >actually technically in p2p/ad-hoc mode... *shrug* it just worked, i didn't have | >to toy with it. | | Thanks for the quick response! When I get it working with my Airport base station, I'll post | an update - I'm still waiting on delivery of my Aibo so it will be a few weeks (also | travelling a bit). | |
| +- 0918 RE:Mac OS X (10.3.2) Panther Support bob_ebert [2004/01/24-07:33] |
| | Response : Email : Edit : Delete |
|
| >I've
just joined and was wondering if anyone that has set up their
development | >environment under Mac OS X (10.3 or earlier) would be kind enough to share | any hints, | >tips, traps, etc.? | | Also just joined, but I didn't have any problems installing the SDK under | Panther. Followed the directions, and the Mac install script worked just | fine. HelloWorld and the other sample programs in the SDK build and work | correctly. I'm bummed the remote processing option is not available. | | >Anyone communicating with their aibo using an Airport terminal? Tips? | | Also use an Airport (original, not Extreme) and had a few problems getting the | ERS-7 to use the connection. No problem with the closed network or WEP key, | that worked perfectly. However, I could not get the SSDP stuff to work, so a | PC on the Airport network was not able to find the ERS-7 using the "AIBO | Search" feature. I ended up turning off DHCP and manually configuring the | robot's IP address, subnet, and gateway. I use 10.0.1.254 so it's unlikely | that the airport will issue a conflicting address. I also had to manually add | the DNS address. | | However, the SDK Console (telnet to port 59000) and the TinyFTPD for Aibo work | flawlessly. You can't use Mac OS's ftp mounting support directly to upload to | the robot, since their built in file system doesn't support writing to FTP | servers, but get a shareware/freeware FTP client or use the command line | version, no problem. The POP/SMTP support in the Aibo Mind software also work | flawlessly with the Airport. | |
| +- 0922 RE:Mac OS X (10.3.2) Panther Support rodrigoa [2004/01/25-04:19] |
| | Response : Edit : Delete |
|
| >>I've
just joined and was wondering if anyone that has set up their
development | >>environment under Mac OS X (10.3 or earlier) would be kind enough to share | >any hints, | >>tips, traps, etc.? | > | >Also just joined, but I didn't have any problems installing the SDK under | >Panther. Followed the directions, and the Mac install script worked just | >fine. HelloWorld and the other sample programs in the SDK build and work | >correctly. I'm bummed the remote processing option is not available. | > | >>Anyone communicating with their aibo using an Airport terminal? Tips? | > | >Also use an Airport (original, not Extreme) and had a few problems getting the | >ERS-7 to use the connection. No problem with the closed network or WEP key, | >that worked perfectly. However, I could not get the SSDP stuff to work, so a | >PC on the Airport network was not able to find the ERS-7 using the "AIBO | >Search" feature. I ended up turning off DHCP and manually configuring the | >robot's IP address, subnet, and gateway. I use 10.0.1.254 so it's unlikely | >that the airport will issue a conflicting address. I also had to manually add | >the DNS address. | > | >However, the SDK Console (telnet to port 59000) and the TinyFTPD for Aibo work | >flawlessly. You can't use Mac OS's ftp mounting support directly to upload to | >the robot, since their built in file system doesn't support writing to FTP | >servers, but get a shareware/freeware FTP client or use the command line | >version, no problem. The POP/SMTP support in the Aibo Mind software also work | >flawlessly with the Airport. | | I'm using an Airport without a Base Station (Peer to Peer) on OSX 10.3.2 and it works | fine. The only trick is that the wireless network will not assume the manualy specified IP | until the AIBO boots up. | |
| +- 0927 RE:Mac OS X (10.3.2) Panther Support sapient [2004/01/26-18:47] |
| | Response : HomePage : Email : Edit : Delete |
|
| I'm
using the airport on a powerbook G4, all is ok, i think everything I've
encountered is | above. | | I was wondering if anyone has any knowlege about using remote "processing OPEN-R" | it seems that this is possible for cygwin or linux but I run into problems (on the Mac) in: | | rp-openr-config | | When trying to make the host side of the program. | | The config file doesn't recognise the operating system. I'm guessing that this means | that OPENR won't run on a mac unless it is recompiled for it. But I thought is was worth | just throwing this question out to see... | | Does Sony have any plan to recompile the source to run under Darwin? | |
| +- 0931 RE:Mac OS X (10.3.2) Panther Support bob_ebert [2004/01/27-09:27] |
| | Response : Email : Edit : Delete |
|
| >
Does Sony have any plan to recompile the source to run under Darwin? | | In the FAQ: A0031 | https://openr.aibo.com/openr/eng/perm/openrsdk/regi_faq2.php4#A0031 | |
| |- 0932 Author DELETE ***** [2004/01/27-18:28] |
| | Response : Edit : Delete |
|
| (The
article that 2004/01/27-18:24 posted, was modified by Author) | |
| +- 0933 RE:Mac OS X (10.3.2) Panther Support sapient [2004/01/27-18:29] |
| Response : HomePage : Email : Edit : Delete |
|
Uh,
sorry didn't check the faq Looking stuppid now :o( (The article that 2004/01/27-18:27 posted, was modified by Author) |
| V 0916 Aibo doesn't want to power down manfred [2004/01/22-01:57] |
| | Response : Email : Edit : Delete |
|
| We
tried to run custom software on an Aibo ERS7, the software didn't work.
When | we tried to shut it off the power light just keeps blinking and nothing happend. | We can't remove the battery because the Aibo won't let us. | | Does anyone know how to solve this? | |
| +- 0920 RE:Aibo doesn't want to power down TodboT [2004/01/24-10:38] |
| | Response : Edit : Delete |
|
| There's
a release to remove the battery and memory stick. Just insert a paper | clip into the hole by the wireless lan switch. Why the software doesn't work, | I'm not sure. You'll have to give some more details. Do the sample programs work? | |
| +- 0929 RE:Aibo doesn't want to power down manfred [2004/01/26-22:32] |
| Response : Edit : Delete |
|
>You'll
have to give some more details. Do the sample programs work? Yes, the sample programs work. We were trying to port the German Team RoboCup software from last year. |
| V 0923 GetTime() Clock() How long does a rutine take rodrigoa [2004/01/25-04:26] |
| | Response : Edit : Delete |
|
| Im
trying to figure out how long does a rutine take to excecute Has
anybody ever used | something like GetTime() or Clock() successfuly? Is there any way to get subsecond | resolution or access the system clock ticks? | (The article that 2004/01/25-04:24 posted, was modified by Author) | |
| +- 0924 RE:GetTime() Clock() How long does a rutine take ejt [2004/01/25-11:07] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Im
trying to figure out how long does a rutine take to excecute Has
anybody ever used | >something like GetTime() or Clock() successfuly? Is there any way to get subsecond | >resolution or access the system clock ticks? | >(The article that 2004/01/25-04:24 posted, was modified by Author) | If you don't mind the license (LGPL) you can yank the handy TimeET utility class out of our | framework... | Header: http://www.tekkotsu.org/dox/TimeET_8h-source.html | Implementation: http://www.tekkotsu.org/dox/TimeET_8cc-source.html | That will theoretically give you microsecond accuracy. Make sure to define PLATFORM_APERIOS if | you want to compile for the Aibo, and undefine it if you want to use it in a unix program (it's | portable) There is one (perhaps fixable) issue - on Aperios TimeET will hold time since boot, and | on a unix system it will hold a real time of day. | | For lighter weight access to millisecond time since boot, we also use a get_time() function | Header: http://www.tekkotsu.org/dox/TimeET_8cc-source.html | Implementation: http://www.tekkotsu.org/dox/TimeET_8cc-source.html | Which again, depending on the definition of PLATFORM_APERIOS, will either return milliseconds | since boot, or will return a global simulator_time variable that you can use to run simulations... | | You might also want to look into using our Profiler itself, since that would seem to directly address | the problem you're trying to solve: | http://www.tekkotsu.org/dox/classProfiler.html | Sample profiling reports here: | http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/docs/benchmarks/ | (The article that 2004/01/25-11:04 posted, was modified by Author) | |
| +- 0925 RE:GetTime() Clock() How long does a rutine take rodrigoa [2004/01/25-12:19] |
| Response : Edit : Delete |
|
>>Im
trying to figure out how long does a rutine take to excecute Has
anybody ever used >>something like GetTime() or Clock() successfuly? Is there any way to get subsecond >>resolution or access the system clock ticks? >>(The article that 2004/01/25-04:24 posted, was modified by Author) >If you don't mind the license (LGPL) you can yank the handy TimeET utility class out of our >framework... >Header: http://www.tekkotsu.org/dox/TimeET_8h-source.html >Implementation: http://www.tekkotsu.org/dox/TimeET_8cc-source.html >That will theoretically give you microsecond accuracy. Make sure to define PLATFORM_APERIOS if >you want to compile for the Aibo, and undefine it if you want to use it in a unix program (it's >portable) There is one (perhaps fixable) issue - on Aperios TimeET will hold time since boot, and >on a unix system it will hold a real time of day. > >For lighter weight access to millisecond time since boot, we also use a get_time() function >Header: http://www.tekkotsu.org/dox/TimeET_8cc-source.html >Implementation: http://www.tekkotsu.org/dox/TimeET_8cc-source.html >Which again, depending on the definition of PLATFORM_APERIOS, will either return milliseconds >since boot, or will return a global simulator_time variable that you can use to run simulations... > >You might also want to look into using our Profiler itself, since that would seem to directly address >the problem you're trying to solve: >http://www.tekkotsu.org/dox/classProfiler.html >Sample profiling reports here: >http://cvs.tekkotsu.org/cgi-bin/viewcvs.cgi/Tekkotsu/docs/benchmarks/ >(The article that 2004/01/25-11:04 posted, was modified by Author) Thanks!!! I'll look into it. |
| V 0912 older Open-R Versions tarx [2004/01/21-09:07] |
| | Response : Email : Edit : Delete |
|
| Hi, | I wanted to start working with an ers-210 with a school class. As I understand | it, the new Version does not support this model anymore. Where can i download | the older version (1.4) ? | | Guido | |
| +- 0919 RE:older Open-R Versions TodboT [2004/01/24-10:30] |
| Response : Edit : Delete |
|
1.4
is also in the download section. |
| V 0917 Tekkotsu Version 2.0 Released ejt [2004/01/22-10:08] |
| Response : HomePage : Email : Edit : Delete |
|
We
are pleased to announce the 2.0 release of the Tekkotsu framework. This is a summary of the changes, but for full detail, please visit: http://www.tekkotsu.org/VersionHistory.html New Features: * Full support for the ERS-7 * New Vision Pipeline - Includes JPEG compression and streaming (allows transmission at full frame rate) * VisionGUI supports dynamic resolution changes, new formats * Sensor Observer - new control allows logging to console or file * Posture Editor - new control allows numeric editing of pose; save and load - a good way to experiment with the new ERS-7 face panel (since there aren't any released diagrams yet) * Event Logger has a verbosity level 3, lets you log Event subclass specific data, such as the current velocity of a LocomotionEvent API Changes: * Changes have been made to the project Makefile and default StartupBehavior_Setup files. Be sure to update these in your projects so you can access the new menus. - a project patch file is available from the Version History, this can help you determine what's changed * WalkMC now has calibration parameters so your desired target velocities can hopefully get within the ballpark of actual achieved motion - An interactive walk calibrator is included under the Walk Edit/calibration menu This will help you log data, but the actual math is done in MATLAB (script included) * Shared/WorldStateSeralizer is now Behaviors/Mon/WorldStateSerializerBehavior - also includes a change in the communication protocol. See Version History for details. * Other small stuff... probably won't affect many, but be sure to see the Version History Bug Fixes: * Improves reaction time by 24ms by using most recent frame of system's sensor update buffer * Other small stuff.... (see Version History) Known Bugs: * VisionGUI has a moderate memory leak when viewing a JPEG stream * A very rare freeze up when deactivating the Emergency Stop * Kinematics have not been updated for the ERS-7. Editing the walk might be unintuitive. * Other small stuff... (see Version History) Documentation: * FAQ: - How can I plug in the power adapter without shutting down? http://www.tekkotsu.org/FAQ.html#power_adapter - How can I dynamically change configuration parameters? http://www.tekkotsu.org/FAQ.html#change_config * New demo video: http://www-2.cs.cmu.edu/~tekkotsu/Samples.html#stabilization * Vision overview: http://www.tekkotsu.org/Vision.html License Change: * Now using the LGPL instead of GPL. Mainly, this makes more sense for a framework where the whole point is to build essentially unrelated things on top of it. Contact info@tekkotsu.org if you really want a detailed explanation for all the reasoning. All contributors were contacted and agreed to the change. |
| V 0900 SDK tutorial #5 corrupt RingRipper [2004/01/18-06:50] |
| | Response : Email : Edit : Delete |
|
| The
subject pretty much says it all. I can't download and save tutorial #5 | without getting a "corrupt file" error message when I try to open it. All the | other tutorials work okay. | | Yes, I have Adobe Reader 6.0.1, and yes I've installed the Japanese fonts. | |
| +- 0913 RE:SDK tutorial #5 corrupt openrsupport@(Administrator) [2004/01/21-13:38] |
| | Response : Edit : Delete |
|
| I
was able to download the SDK tutorial #5 just fine. | What error message do you get when you try to download tutorial #5? | What web browser are you using? | Did you clear the cashed file and try again? | | Thank you, | | OPEN-R SDK Support | |
| +- 0915 RE:SDK tutorial #5 corrupt RingRipper [2004/01/21-20:54] |
| Response : Email : Edit : Delete |
|
Mine
downloads fine, too. It's when you go to open it that you get an error message stating that the file is corrupt. I tried again -- the fifth or sixth time -- and it finally worked. |
| V 0906 how to programm the Model of ERS-7 totoduke [2004/01/20-02:22] |
| | Response : Edit : Delete |
|
| Hi, | | I am now using the ERS-7 for my research. I now have great difficulties in | programming the Robot because I don't know how I can get the parameters of | this model. Are there many differences between the model ERS-2xx and ERS-7? I | am now trying to write a walking code which seems to be very difficult for one | who have not much experience on programming using C++? | | Can someone offer some helps here? Thanks a lot. | |
| +- 0908 RE:how to programm the Model of ERS-7 AiboPet [2004/01/20-03:52] |
| | Response : Edit : Delete |
|
| >I now have great difficulties in | >programming the Robot because I don't know how I can get the parameters of | >this model. | | Check the Download page. There is an updated model reference for the ERS-7 on | there now. It includes dimensions, range of angles etc. | | > Are there many differences between the model ERS-2xx and ERS-7? | | Depending on your walking algorithm, the differences may or may not matter. | The big difference I found is in the center of gravity (more important for | skits with the robot sitting down and trying to move his arms). The neck | structure is significantly different. | | Legs are relatively the same. I took the CMU walking code (using the knee | walk) originally written for the ERS-210. Just by renaming the servos, it | works well on the ERS-7. | | That is used in my AiboSpeed Racer program. | http://www.aibohack.com/openr_sdk/speedracer.htm | | >am now trying to write a walking code which seems to be very difficult for one | >who have not much experience on programming using C++? | | I do not recommend writing a walking routine from scratch (unless that is the | purpose of your research). | | For the ERS-7 (or ERS-2x0), the CMU code provides a very precise and fast knee | walk. The original CMU Pack 02 code or my derived version should get your AIBO | walking without a lot of work. | | If you want a more normal walk (ie. not on the knees) there is a library for | regular walks called the "OMWares" library, which is used in the SoccerLion | example. It is available in binary form only from Sony. It currently only | supports the ERS-2x0. I am eagerly waiting for an ERS-7 compatible version. | |
| +- 0910 RE:how to programm the Model of ERS-7 totoduke [2004/01/20-04:39] |
| | Response : Edit : Delete |
|
| >>I
now have great difficulties in | >>programming the Robot because I don't know how I can get the parameters of | >>this model. | > | >Check the Download page. There is an updated model reference for the ERS-7 on | >there now. It includes dimensions, range of angles etc. | > | >> Are there many differences between the model ERS-2xx and ERS-7? | > | >Depending on your walking algorithm, the differences may or may not matter. | >The big difference I found is in the center of gravity (more important for | >skits with the robot sitting down and trying to move his arms). The neck | >structure is significantly different. | > | >Legs are relatively the same. I took the CMU walking code (using the knee | >walk) originally written for the ERS-210. Just by renaming the servos, it | >works well on the ERS-7. | > | >That is used in my AiboSpeed Racer program. | >http://www.aibohack.com/openr_sdk/speedracer.htm | > | >>am now trying to write a walking code which seems to be very difficult for | one | >>who have not much experience on programming using C++? | > | >I do not recommend writing a walking routine from scratch (unless that is the | >purpose of your research). | > | >For the ERS-7 (or ERS-2x0), the CMU code provides a very precise and fast knee | >walk. The original CMU Pack 02 code or my derived version should get your AIBO | >walking without a lot of work. | > | >If you want a more normal walk (ie. not on the knees) there is a library for | >regular walks called the "OMWares" library, which is used in the SoccerLion | >example. It is available in binary form only from Sony. It currently only | >supports the ERS-2x0. I am eagerly waiting for an ERS-7 compatible version. | |
| +- 0911 RE:how to programm the Model of ERS-7 totoduke [2004/01/20-04:42] |
| Response : Edit : Delete |
|
>>>I
now have great difficulties in >>>programming the Robot because I don't know how I can get the parameters of >>>this model. >> >>Check the Download page. There is an updated model reference for the ERS-7 on >>there now. It includes dimensions, range of angles etc. >> >>> Are there many differences between the model ERS-2xx and ERS-7? >> >>Depending on your walking algorithm, the differences may or may not matter. >>The big difference I found is in the center of gravity (more important for >>skits with the robot sitting down and trying to move his arms). The neck >>structure is significantly different. >> >>Legs are relatively the same. I took the CMU walking code (using the knee >>walk) originally written for the ERS-210. Just by renaming the servos, it >>works well on the ERS-7. >> >>That is used in my AiboSpeed Racer program. >>http://www.aibohack.com/openr_sdk/speedracer.htm >> >>>am now trying to write a walking code which seems to be very difficult for >one >>>who have not much experience on programming using C++? >> >>I do not recommend writing a walking routine from scratch (unless that is the >>purpose of your research). >> >>For the ERS-7 (or ERS-2x0), the CMU code provides a very precise and fast knee >>walk. The original CMU Pack 02 code or my derived version should get your AIBO >>walking without a lot of work. >> >>If you want a more normal walk (ie. not on the knees) there is a library for >>regular walks called the "OMWares" library, which is used in the SoccerLion >>example. It is available in binary form only from Sony. It currently only >>supports the ERS-2x0. I am eagerly waiting for an ERS-7 compatible version. Thank you very much. The basic part of My research is about walking. Actually, I only some simple normal walk function which can make the dog walk in a desired path. |
| V 0902 AiboMind won't reboot anymore. How to restore backup? angel [2004/01/18-21:21] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
got a big problem. I tried to install AiboMindPlus, supposely an
addition you | can install on top of AiboMind in order to monitor telemetry from the ERS-7. | Well I did a backup of the memory stick first and then did the installation. I | don't think it work well. Then I wanted to restore the original AiboMind | software, so I remove the OPEN-R folder and restore my backup from the PC. Now | the ERS-7 will not reboot anymore. | | I went to the FAQ on this website and notice that AiboMind is not extensible | period, which makes me now wonder how the AiboMindPlus could then work in the | first place. | | Anyway how, my really big problem is that I have a backup of the AiboMind memory | stick and I can't make it work. Please any help?? | |
| +- 0903 RE:AiboMind won't reboot anymore. How to restore backup? AiboPet [2004/01/19-02:57] |
| | Response : Edit : Delete |
|
| AiboMindPlus
is an unofficial addition (everything on my site is "USE AT YOUR | OWN RISK"). It has nothing to do with the OpenR SDK. | | I suggest posting non-OpenR SDK questions on the public BBSs such as: | http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi?ubb=forum;f=5 | | What brand memory stick reader are you using ? | ---- | | That said, the common problems, especially for the first time use of a stick | reader are: | | 1) you are not using a Sony memory stick reader, and your stick reader broke | the copy protection [bad] | 2) your drivers aren't working properly and the FAT format is screwed up | [usually fixable] | 3) user error - you are restoring to the wrong folder (eg: OPEN-R\OPEN-R) | [fixable] | | > ...Then I wanted to restore the original AiboMind | >software, so I remove the OPEN-R folder and restore my backup from the PC. Now | >the ERS-7 will not reboot anymore. | | First, be sure you are using a Sony brand stick reader ! | | Check the stick and be sure there is an "OPEN-R" folder at the top, | and "APP", "MW" and "SYSTEM" inside that folder. | |
| +- 0905 HOWTO LAN Console on ERS-7 angel [2004/01/19-23:08] |
| | Response : HomePage : Email : Edit : Delete |
|
| Thanks
for the help. But I figured it out, it was simple. | | I am using a Sony VAIO Laptop with a Sony reader. So that wasn't the problem. | Since I left the Aibo out of its charging station for a week or two, even though | I didn't use it, still the battery discharged because (as the manual clearly | indicates) there is no stop button. Aibo is always using the battery unless on | the recharger. So essentially the problem was that the battery was completely | drained. It was a coincidence that the last juice left on it dissappeared right | after I installed the AiboMind Plus addition. That was what gave me the wrong | impression that it was something wrong with the installation. Later on my backup | restores of the original AiboMind just appear not to work because there was not | charge in the battery at all. | | It is too bad that AiboMind doesn't come with something like AiboMindPlus and | Scope7. In the last two days I have learned more about how Aibo works (and why | it doesn't) with that simple addition than since Decemeber when I bought it. The | voice recognition is pretty bad, Aibo is misinterpreting all the time what it | hears. Another problem it has is a more difficult to deal with, it doesn't now | how to tune out background conversations and joint noise from actual commands | issued to him. The camera is also a bit pathetic since it needs a great luminance. | | The sound recognition is something that a programmer/owners of Aibo can try to | fix and experiment a bit more. The limitations of the camera are more serious. | You can only work around that by installing a photo studio in you house and | putting mega-watts of Lumens around your living room. Just kidding.... :) | | >AiboMindPlus is an unofficial addition (everything on my site is "USE AT YOUR | >OWN RISK"). It has nothing to do with the OpenR SDK. | |
| +- 0909 RE: voice recog AiboPet [2004/01/20-03:56] |
| Response : Edit : Delete |
|
>It
is too bad that AiboMind doesn't come with something like AiboMindPlus
and >Scope7. That's why it is free (for those who want to view their pet as more like a robot). One of the most popular and easy to use features [ie. sit back and watch] > The >voice recognition is pretty bad, Aibo is misinterpreting all the time what it >hears. The trick is to train *yourself* to be better at giving the commands. Sometimes humans are more adaptable than robots ;-> Background noise is (anecdotally) more of an issue with the ERS-7. I suspect it may be because they simplified the voice command list (compared to AiboLife2). |
| V 0858 BlinkingLED and Tail LEDs of ERS-210 lukas_hohl [2003/12/18-07:34] |
| | Response : Email : Edit : Delete |
|
| I'm
trying to extend the functionality of the BlinkingLED sample to the
Tail LEDs of an ERS-210. | There's no error | when opening the corresponding Primitives. However, the Tail LEDs do not react to commands | similar to those given to the Head LEDs in the BlinkingLED sample. Is there any undocumented | special value that I have to put into the command vector? Thanks in advance for your help. | (The article that 2003/12/18-07:33 posted, was modified by Author) | |
| +- 0891 RE:BlinkingLED and Tail LEDs of ERS-210 openrsupport@(Administrator) [2004/01/14-03:33] |
| | Response : Edit : Delete |
|
| Hello, | | There is no undocumented special command. With a very minor change of the CPC | locators, you can blink the tail LEDs. | | $ diff BlinkLED.h BlinkLED.h.org | 26,27c26,27 | < "PRM:/r6/l1-LED2:l1", | < "PRM:/r6/l2-LED2:l2", | --- | > "PRM:/r1/c1/c2/c3/l1-LED2:l1", | > "PRM:/r1/c1/c2/c3/l2-LED2:l2", | | If you are adding more LEDs, you'll also need to update NUM_LEDS. | | OPEN-R SDK Support | |
| +- 0907 RE:BlinkingLED and Tail LEDs of ERS-210 lukas_hohl [2004/01/20-03:07] |
| Response : Email : Edit : Delete |
|
Thanks
for your answer. I did exactly as you proposed (in fact I did again what I had tried before writing my first message) but there was no effect. All but the tail LEDs are blinking. There's no error in the method OpenPrimitives(). Is there a way (other than opening the primitive) to know whether the tail LEDs are working at all? |
| V 0901 Unsable Memory Stick for dog (but still can r/w data) cklam [2004/01/18-18:34] |
| | Response : Edit : Delete |
|
| Some
of my pink ERA-MS016 memory sticks (the most normal one with | 16MB) cannot be used to boot the AIBO dog. If I use some new | memory sticks then everything is fine. | | These unworkable memory sticks are quite old (I've used them for more | than 2 years). Although they can't boot the dog, they can still | store data like the good ones. I wonder if there is any way to | "clean" them somehow to make them usable for dogs again? I tried rubbing | the contacts with paper but that doesn't seem to work. | | Any help will be appreciated, thanks!! | |
| +- 0904 RE:Unsable Memory Stick for dog (but still can r/w data) AiboPet [2004/01/19-03:03] |
| Response : Edit : Delete |
|
>Some
of my pink ERA-MS016 memory sticks (the most normal one with >16MB) cannot be used to boot the AIBO dog. If I use some new >memory sticks then everything is fine. >These unworkable memory sticks are quite old (I've used them for more >than 2 years). If they are very old RoboCup sticks, they won't work on the ERS-210/220 or ERS- 7, period. If they used to work on the ERS-210/220/7 and just stopped working, then I suspect that the copy protection has become corrupted or you reformatted them (and the FAT format is not compatible with AIBO). The former is not user fixable, the latter problem can be fixed with a WinNT/2K/XP machine full format. What brand stick reader are you using ? Have you ever formatted the pink sticks ? (conditional or unconditional format?) [Historical note: I have seen a case where non-Sony brand stick readers work for some time, but eventually corrupt the stick - what I used to call "pink stick rot". Now I only recommend Sony brand stick readers, and it reduced the problems drastically] Assuming you can access them as a regular stick, then the memory stick is still ok. Cleaning the contacts won't change anything. |
| V 0895 It is pink, has no ERF code and it does not work ... dobrisek [2004/01/14-23:01] |
| | Response : Edit : Delete |
|
| We proudly own ESF-7 and: | Sony MSAC-US2 MS Reader Writer, | An extra memory stick ERA-MS016 (brand new without any front label) | WLAN WAP that works fine with AIBO Mind software and AiboScope7_01. | (The same working WLANCONF.TXT was used for OPEN-R examples) | | We biult HelloWorld example from OPEN_R_SDK-sample-1.1.5-beta1.tgz | using OPEN_R_SDK-1.1.5-beta1.tgz on Cygwin/Windows XP environment | (downloaded from this site). We copied MS_ERS7/WCONSOLE/memprot/OPEN-R | to the mentioned empty pink MS. We copied MS/OPEN-R files that | appeared after 'make install' of the example to | the same stick (overwriting OBJECT.CFG and updating WLANCONF.TXT that | works fine with the AIBO Mind software) following Programmer's Guide. | We put MS (with the proper file stucture) into ESF-7 (WLAN is switched on) | and after initial pause botton beep we got some error beeps after 10 secons | and ESF-7 shuts down. The same happened with the BASIC system and BlinkingLED7 | example. | | What we did wrong? Please help! | | HelloWorld example file structure of the pink MS without ERF label !!! ;) | looks like this (with a help of 'tree /F' command) : | | H:. | └───OPEN-R | │ APERINIT.GZ | │ APERIOS.GZ | │ BOOTPARA | │ | ├───SYSTEM | │ │ VERSION | │ │ | │ ├───OBJS | │ │ ANTTCPIO.BIN | │ │ DR.BIN | │ │ EMGCYMON.BIN | │ │ HOOK.BIN | │ │ HOOKACT.BIN | │ │ IPSTACK.BIN | │ │ NETCONFS.BIN | │ │ POWERMGR.BIN | │ │ SRVCMGR.BIN | │ │ SYSLOG.BIN | │ │ VR.BIN | │ │ VRAUCOMM.BIN | │ │ VRCOMM.BIN | │ │ WLANDRV.BIN | │ │ WLANENBL.BIN | │ │ | │ ├───DATA | │ │ ├───P | │ │ │ AIBO-ID | │ │ │ | │ │ └───E | │ │ MSWATCH.DAT | │ │ | │ └───CONF | │ CARDDRV.CFG | │ EMON.CFG | │ EXTOBJ.CFG | │ OBJECT.CFG | │ WLANCONF.TXT | │ | └───MW | ├───OBJS | │ HELLO.BIN | │ POWERMON.BIN | │ | ├───DATA | │ ├───P | │ └───E | └───CONF | CONNECT.CFG | DESIGN.CFG | DESIGNDB.CFG | OBJECT.CFG | | | |
| +- 0897 RE:It is pink, has no ERF code and it does not work ... AiboPet [2004/01/15-02:05] |
| | Response : Edit : Delete |
|
| >We proudly own ESF-7 and: | | I assume you mean "ERS-7" | What you explain sounds like you are doing the right thing. | | I would suggest trying it again, but use the BASIC, nomemprot version first. | Don't worry about the LAN/WLANCONF.TXT right now. | | Install just the MS_ERS7\BASIC\nomemprot version, no sample. Boot and make | sure AIBO doesn't stop after 10 seconds [he won't do anything, and after | waiting 1 minute if he is still running, just turn him off - test passed] | | Then install something that shows feedback, like the blinking LED sample, and | try again (the HelloWorld sample doesn't do anything visible) | | If all this is working, then that proves your installation is fine and the | pink PMS stick is fine. Then try the WCONSOLE version again. | | ------ | | >HelloWorld example file structure of the pink MS without ERF label !!! ;) | >looks like this (with a help of 'tree /F' command) : | >... | > $B('(!(!(!(BDATA | > $B("(B $B('(!(!(!(BP | > $B("(B $B(&(!(!(!(BE | > $B(&(!(!(!(BCONF | > CONNECT.CFG | > DESIGN.CFG | > DESIGNDB.CFG | > OBJECT.CFG | > | | The end of your listing looks incorrect, but it could be a format problem. | Check to make sure these 4 files are in the \OPEN-R\MW\CONF folder and not in | the root. | |
| +- 0898 RE:It is pink, has no ERF code and it does not work ... dobrisek [2004/01/15-22:28] |
| Response : Edit : Delete |
|
>I
assume you mean "ERS-7" Yes, sorry, I mistyped the name. >Install just the MS_ERS7\BASIC\nomemprot version, no sample. Boot and make >sure AIBO doesn't stop after 10 seconds [he won't do anything, and after >waiting 1 minute if he is still running, just turn him off - test passed] This test passed! >Then install something that shows feedback, like the blinking LED sample, and >try again (the HelloWorld sample doesn't do anything visible) Hmmm... this makes ERS-7 to shutdown with the error beeps again. I don't see anything strange in the 'make' log and 'make install' log. However, it is true that I am currently running OPEN-R on my updated Cygwin/Windows XP system (I don't want to overwrite it with older version that can be downloaded from this site and switching between two different complete installations requires some hacking in the system registry) and I am using gcc (GCC) 3.3.1 (cygming special)... hmmm ... yes (!), this was the problem... one of our students tried the same examples with the proper Cygwin installation some days ago and there were the same problems. However, today everything works well. I don't know why this happend for the first time with the original Cygwin setup. Perhaps he forgot to run 'make install' and there was no BINS there, just CFGS... Thanks a lot! |
| V 0854 probs with remote processing GingerJake [2003/12/17-00:46] |
| | Response : Edit : Delete |
|
| Hi all, | Am new to all this so please bear with me. | I am currently installing Open-R SDK on a Linux platform. Have managed to run | the HelloWorld sample so have now moved on to the ObjectComm, have followed the | instructions, but when I get to run start-rp-openr command I get : | | [pid:3097,msquid:65538,oid:0x00010002] oserviceManager | [pid:3098,msquid:98307,oid:0x00018003] tcpGateway | MS/OPEN-R/MW/CONF/OBJECT.CFG: no such file or directory | | The file is definetly there with the correct permissions, I can telnet to the | dog and the program performs as it says it should. | | Any suggestions, probably something really basic that I've completely overlooked | Thanks for your help | Julie | |
| +- 0890 RE:probs with remote processing openrsupport@(Administrator) [2004/01/14-03:13] |
| | Response : Edit : Delete |
|
| Hi Julie, | | We apologize for the very long delay in a response to your problem. We have | been away for the holidays and occupied with other events. | | Please follow the installation guide step by step to run the HelloWorld sample. | | MS/OPEN-R/MW/CONF/OBJECT.CFG is needed on the PC side to run Remote Processing | OPEN-R. | | You should run the start-rp-openr command in the appropriate directory. | | Type | $ls MS/OPEN-R/MW/CONF/OBJECT.CFG | to verify that OBJECT.CFG is in your current folder. | | Thank you, | | OPEN-R SDK Support | |
| +- 0896 RE:probs with remote processing GingerJake [2004/01/15-00:33] |
| Response : Edit : Delete |
|
Hi, Thanks very much for your help, not sure how or why but I have it running now. Thanks again Julie >Hi Julie, > >We apologize for the very long delay in a response to your problem. We have >been away for the holidays and occupied with other events. > >Please follow the installation guide step by step to run the HelloWorld sample. > >MS/OPEN-R/MW/CONF/OBJECT.CFG is needed on the PC side to run Remote Processing >OPEN-R. > >You should run the start-rp-openr command in the appropriate directory. > >Type >$ls MS/OPEN-R/MW/CONF/OBJECT.CFG >to verify that OBJECT.CFG is in your current folder. > >Thank you, > >OPEN-R SDK Support |
| V 0875 SDK 1.1.5beta bugs? ejt [2004/01/08-12:14] |
| | Response : HomePage : Email : Edit : Delete |
|
| There's
two unrelated bugs. One is the OFbkImage::FieldCounter() function. It | always returns 0, not the camera image count I would expect (from page 61 of the | Level 2 Reference) | | However, the following code for extracting the count directly from the last line | of the image seems | to work just fine: | const OFbkImage img(...); | frameNumber=*(int*)(img.Pointer()+(img.Height()-1)*(img.Skip()+img.Width())); | | I haven't tried FieldCounter() on the 2xx series - perhaps it never worked and | I'm just noticing it now. | | (On a side note, storing this information in the image itself is rather annoying | since this will confuse some vision algorithms since they think there's | something there. Either I have to skip using the last row (which is a pain | because now images have different dimensions) or I have to overwrite the | affected bytes with the row above (which has problems of its own). But I'm | guessing that this is how the hardware stores the information, so we're stuck | with it? Otherwise, it's be much more straightforward to just put the | information in some fields in OFbkImage itself instead of overwriting image | pixels...) | | The other bug is with the PID control. It may not be a bug, but it's different | from the 2xx series. I've found I need to limit the gain parameters to 7 bits | of precision (signed byte I guess), and avoid using 0 shifts. Otherwise joints | go out of control. | Being able to control and understand the PID values is important for | manipulation tasks, where I care about the force being applied as much as the | current position. Since my Masters thesis is going to involve manipulation | tasks, it would be really nice if some documentation could be provided on the | PID parameters and how they are used to compute the duty cycles. | As long as I'm making requests, it would be even better if you could just | provide a straightforward interface that takes a single floating point value for | each of p,i, and d, instead of messing around with this odd gain/shift interface. | | Also, even with the constraints I listed above, for some reason there's a | significant twitch that occurs during boot. What I'm doing is fading in the PID | values linearly from 0 to the default values | over the course of 2 seconds, at 128 ms increments. I do this so that there | isn't a sudden twitch when the motors adjust from the sensed position to the | driven position that corresponds to (since there's often a bit of miscalibration | between the sensors and motors). Fading in the PIDs allows it to gradually | tighten up instead. | | So, for instance, for the nod joint of the ERS-7, the pids look like (in decimal): | P I D | G S G S G S | 16 2 0 2 8 2 | 35 7 0 2 35 6 | 44 8 0 2 44 7 | 38 9 0 2 38 8 | 55 9 0 2 55 8 | 36 10 1 2 36 9 | 44 10 1 2 44 9 | 53 10 1 2 53 9 | 62 10 1 2 62 9 | 34 11 2 2 34 10 | 38 11 2 2 38 10 | 42 11 2 2 42 10 | 46 11 2 2 46 10 | 51 11 3 2 51 10 | 55 11 3 2 55 10 | 58 11 3 2 58 10 | 63 11 3 2 63 10 | 4 15 2 3 2 15 <= twitch occurs when making the jump to the final value | (sigh, this BB kills my tabular formatting...) | | This works fine on the 2xx series, but that twitch on the ERS-7 worries me that | there's something strange going on. | | Finally, with the new limitations on the gains and shifts, i'm getting a lot of | warnings on the 210 | that, although I had gotten before, I had managed to avoid by fading in over | 2047 ms instead of just 2000 ms. (the prime number caused the gain to stay at | high precision the whole time until the final setting) Now I see this a lot | during the inital fade in (but everything works fine) | [oid:80000035,prio:1] AGRMSDriver::SetGain() : 0x0A IS USED FOR GAIN SHIFT VALUE. | | I have no idea why 0x0A would be an invalid shift value - it seems to work fine, | and the defaults have 0x0E and 0x02, so it isn't too high or to low...? | | thanks! | (The article that 2004/01/08-12:11 posted, was modified by Author) | |
| +- 0876 RE:SDK 1.1.5beta bugs? (PS) ejt [2004/01/08-12:27] |
| | Response : HomePage : Email : Edit : Delete |
|
| The
code which I'm using to make the conversions from any given floating
point | P, I, or D value to the corresponding gain/shift pair in the updatePIDs function: | http://tekkotsu.no-ip.org/dox/MotionManager_8cc-source.html#l00412 | | Note that the constraints that are actually in this code are more conservative | than I found to be really necessary - that helps reduce the twitch (which can be | bad enough to trigger a "jam condition" shutdown) | | Also, if you use the commented section that enforces the default shifts, the | problem will go away. However, that's pretty limiting, especially for the 'I' | parameter, since the default value is only 1, there's no way to lower it without | changing the shift. | |
| +- 0893 RE:SDK 1.1.5beta bugs? (PS) openrsupport@(Administrator) [2004/01/14-03:59] |
| | Response : Edit : Delete |
|
| Hello, | | Thank you for the bug report. | | 1. OFbkImage::FieldCounter() function always returns 0. | | We acknowledge the bug. It is being analyzed. | | 2. PID control. | | See the Model Information document | (OPEN_R_SDK-docJE-1.1.5-beta1_ModelInformation_7). | | In section 2.3 | (Servo Gain) | Use standard values normally, and use weak gain values when | vibration is hard. PSHIFT, ISHIFT, DSHIFT are fixed values | and do not change the values. | | [Technical comments] | The ERS-200 series hardware can set SHIFT values for each joint. | The ERS-7 has only one register to set the SHIFT value. | All of the joints use same SHIFT value. | Anyway, it is better not to change the SHIFT values. | | Thank you, | | OPEN-R SDK Support | |
| +- 0894 RE:SDK 1.1.5beta bugs? (PS) ejt [2004/01/14-14:33] |
| Response : HomePage : Email : Edit : Delete |
|
>
[Technical comments] > The ERS-200 series hardware can set SHIFT values for each joint. > The ERS-7 has only one register to set the SHIFT value. > All of the joints use same SHIFT value. Ohhhh! :) no wonder! > Anyway, it is better not to change the SHIFT values. Yes, this would make sense given the shared SHIFTs. However, if I may make a suggestion regarding the ERS-7's default PIDs (and please comment on it if there's something amiss), it would seem better to decrease the standard D-shift by at least one or two (but not much more I think), and then increase the standard D-gains accordingly. This will retain the same resultant setting for the actual hardware, but will allow better precision for in-between values. You may notice that most joints currently have a standard D-gain of 1, which means if you want to decrease it, the only option is to set it to 0, there's nothing in between. Decreasing the D-shift by two would mean a standard gain of 4 (for joints normally with a gain of 1), which would allow a little more flexibility to lower it if desired, but still plenty of room to increase the upper end (which is currently 4, would instead be 16 aka 0x10) thanks for the answer, i know this is pretty arcane. Also, if you could provide the formula that is used to determine the duty cycle from the current state, that would be helpful. I've seen a few variants of PID control around, it's hard to model the force being applied to an object without knowing which variant is in use. To do it right, I'd also need to know the max torques of the different motors being used, if this information is can be released... :) I suppose I could always try measure it myself otherwise.... thanks! -ethan |
| V 0860 Head Sensors of ERS-210 lukas_hohl [2003/12/18-07:42] |
| | Response : Email : Edit : Delete |
|
| The
two touch sensors on the head of the ERS-210 return 980665 as the
maximum value. This is | exactly one third of the value 2941995 (= 2.941995N = 300gf) which is indicated in the ERS-210 | model information. Which one is correct? Do I have to multiply the values read from the sensors by | three to obtain the correct value in micro Newton? Thank you for your help. | |
| +- 0892 RE:Head Sensors of ERS-210 openrsupport@(Administrator) [2004/01/14-09:51] |
| Response : Edit : Delete |
|
Thank
you for your query. We are regretful that this is indeed an error in our specification. The actual figure is 980,665 as you suggest. OPEN-R SDK 1.1.4 (for ERS-200 series) documents have been released. The error has been fixed in the new document. Thank you, OPEN-R SDK Support (The article that 2004/01/14-09:50 posted, was modified by Author) |
| V 0865 Opening angles of the ERS-7 camera GTleader [2003/12/31-09:47] |
| | Response : HomePage : Email : Edit : Delete |
|
| The
documentation for the ERS-7 lacks a lot of information that is available | for the ERS-210, e.g. the opening angles of its camera or the measuring ranges | of the PSD sensors. When will this information be available? | |
| +- 0873 RE:Opening angles of the ERS-7 camera openrsupport@(Administrator) [2004/01/06-15:23] |
| | Response : Edit : Delete |
|
| Hello, | | Please see FAQ ID:E0011. | Q. | What is the distribution schedule for the OPEN-R SDK for | ERS-7? | A. | The master release of OPEN-R SDK for ERS-7 will be February, | 2004. When the master SDK is released, the PDF manual including | model data will be released, and sample programs will be added. | | But we realize you may need this data sooner rather than later. | We will do our best to give you this information in a timely manner. | Please understand it will take some time to prepare. | | FYI, the opening angles of ERS-7's camera is 56.9(H)x45.2(V)[deg] | | We will provide the measuring ranges of the PSD sensors as soon | as possible. | | Thank you, | | OPEN-R SDK Support | |
| +- 0889 RE:Opening angles of the ERS-7 camera openrsupport@(Administrator) [2004/01/14-02:55] |
| Response : Edit : Delete |
|
Hello, Here are the measuring distances of the PSD sensors: "PRM:/p1-Sensor:p1", // BODY PSD value 100000 - 900000 (10cm - 90cm) (Note) The BODY PSD doesn't directly measure the distance from the sensor to the floor. It is used to detect cliffs in AIBO MIND. Be careful when using the BODY PSD for other purposes. "PRM:/r1/c1/c2/c3/p1-Sensor:p1", // HEAD PSD (NEAR) value 50000 - 500000 (5cm - 50cm) "PRM:/r1/c1/c2/c3/p2-Sensor:p2", // HEAD PSD (FAR) value 200000 - 1500000 (20cm - 150cm) Thank you, OPEN-R SDK Support |
| V 0888 Aibo joints speed limit aung [2004/01/13-22:01] |
| Response : Email : Edit : Delete |
|
Hi..
all where to check the speed limits of Aibo joints. I know they are somewhere in Sony web site but I just can't remember it. Pls help me asap. |
| V 0884 ERS-7 Mode A/B LEDs ejt [2004/01/11-07:05] |
| Response : HomePage : Email : Edit : Delete |
|
Just
thought I should point out that although it appears that we can set individual LEDs to be mode A or B, so that there can be a mix of both at any given time, my experiments have shown that only the mode of the last LED given is used for ALL of the LEDs. This means that all the LEDs must share the same mode. If this is incorrect or may change in the future, I'd appreciate being corrected since I'm planning on having a single global common setting in my own code. On a side note, why introduce the complication of A vs. B? Especially since for most of the LEDs, the two settngs are identical. Why not just allow direct access to a few more LEDs? Having a mode which selects between them makes it more complicated, and less flexible (since both can't be on at once). |
| V 0883 Error message with sample/ObjectComm angel [2004/01/11-00:24] |
| Response : HomePage : Email : Edit : Delete |
|
I
just install the ObjComm sample of the SDK on my ERS-7 and I get an
error that I don't kind understand. When running the start-rp-openr I get this: $ ~/OPEN_R_SDK/RP_OPEN_R/bin/start-rp-openr [pid:4032,msqid:852,oid:0x00000354] oserviceManager [pid:3976,msqid:853,oid:0x00000355] tcpGateway [pid:4048,msqid:854,oid:0x00000356] MS/OPEN-R/MW/OBJS/SUBJECT.BIN [pid:4088,msqid:855,oid:0x00000357] MS/OPEN-R/MW/OBJS/OBSERVER.BIN TCPGW: Error reading MS/OPEN-R/MW/CONF/HOSTGW.CFG SampleSubject::Ready() : ASSERT READY SampleObserver::Notify() !!! Hello world !!! SampleSubject::Ready() : ASSERT READY SampleObserver::Notify() !!! Hello world again !!! SampleSubject::Ready() : ASSERT READY I reports of an error reading a HOSTCW.CFG file. As a matter of fact, that file doesn't exist before or after the "make install". Nevertheless from the log messages it appears that the observer subjects are communicating. But here is my second problem. When Aibo is turn OFF and I run the start script I get exactly the same messages. How is that possible? |
| V 0862 rp-openr? bigjingo [2003/12/18-21:51] |
| | Response : Email : Edit : Delete |
|
| I
have installed OPEN_R_SDK-1.1.3 in my linux system. However, when I run
the | program setup-rp-openr some errors occurred. All 4 objects come to the same error: | | /usr/local/OPEN_R_SDK/bin/mipsel-linux-ld: cannot open crt1.o: No such file or | directory | collect2: ld returned 1 exit status | | What's wrong with this? How can I solve the problem? | Thanks a lot for your help. | |
| +- 0874 RE:rp-openr? openrsupport@(Administrator) [2004/01/06-15:36] |
| Response : Edit : Delete |
|
>I
have installed OPEN_R_SDK-1.1.3 in my linux system. However, when I run
the >program setup-rp-openr some errors occurred. All 4 objects come to the same error: > >/usr/local/OPEN_R_SDK/bin/mipsel-linux-ld: cannot open crt1.o: No such file or >directory >collect2: ld returned 1 exit status > >What's wrong with this? How can I solve the problem? >Thanks a lot for your help. Hello, First, please make sure that you installed the latest SDK (OPEN_R_SDK-1.1.4). For Remote Processing OPEN-R, the x86 compiler (gcc) should be used instead of mipsel-linux-gcc. According to your error message, mipsel-linux-ld is being used. Did you set the CXX environment to use mipsel-linux-gcc (the mips-compiler) as the default compiler? Try unsetting the CXX environment variable and use the normal gcc compiler (default compiler in your linux system) when you use RP OPEN-R. You don't need to set CXX even if you compile mips objects for AIBO. See FAQ A0036 and D0007. Thanks, OPEN-R SDK Support |
| V 0872 Wireless LAN and ERS-7 gijoe [2004/01/06-12:27] |
| Response : HomePage : Email : Edit : Delete |
|
Hi ! We are having a problem porting our ERS-210 code to the ERS-7. Our TCP/IP code for sending data from a robot to a PC seems to leak memory. I haven't checked exactly, but I believe that every byte that we send over the wireless network is leaked. This is quite perplexing since there is no explicit memory management code anywhere near our TCP/IP code. Apart from the different camera image size, this code has not changed at all during the port. We have had no other problems in porting to the ERS-7. Has anyone had a similar issue ? -Craig. |
| V 0867 HOWTO LAN Console on ERS-7 angel [2003/12/31-23:03] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
have a simple question. My ERS-7 already has the wireless setup and
also the | email facility configured. Now I am getting started with the OpenR programming | guide. The first program I want to try is the HelloWorld example. Unfortunately | I am completly puzzled as to what LAN console the program (old) documentation | keeps referring to. As far as I know I have no LAN console on my PC. I try | finding something on the AIBO LAN Manager (included on the bot CD) but doesnt | seem to be there either. Any hints? THanks | |
| +- 0868 RE:HOWTO LAN Console on ERS-7 AiboPet [2004/01/01-03:05] |
| | Response : Edit : Delete |
|
| >I
have a simple question. My ERS-7 already has the wireless setup and
also the | >email facility configured. | | I assume you mean the feature of the AiboMind software (AiboEyes/AiboMail) | That has nothing to do with OpenR programming, but it is a good test to see | your PC WiFi is working. | | > Now I am getting started with the OpenR programming | >guide. The first program I want to try is the HelloWorld example. | | I assume you are using a different memory stick for this (a PMS/RCode stick, | the 32MB AiboMind stick will not work) | | >Unfortunately | >I am completly puzzled as to what LAN console the program (old) documentation | | This refers to a 'telnet' session you establish between your PC and you AIBO, | usually early on in the boot sequence. | | See the installation guide where it talks about 'telnet' | | The built-in Windows telnet will do (or HyperTerminal will also work). Connect | to port 59000 for the 'console' | For the normal IP address: | telnet 10.0.1.100 59000 | | >I try | >finding something on the AIBO LAN Manager (included on the bot CD) | | That's for the AiboMind software, I don't recommend you use it. | | [it can be used to help configure the WLANCONF.TXT, but IMHO if you are | programming at the OpenR SDK level, you should be able to edit the | WLANCONF.TXT file yourself] | |
| +- 0869 HOWTO LAN Console on ERS-7 angel [2004/01/01-03:49] |
| | Response : HomePage : Email : Edit : Delete |
|
| >>I
have a simple question. My ERS-7 already has the wireless setup and
also the | >>email facility configured. | > | >I assume you mean the feature of the AiboMind software (AiboEyes/AiboMail) | >That has nothing to do with OpenR programming, but it is a good test to see | >your PC WiFi is working. | | Exactly. | | >I assume you are using a different memory stick for this (a PMS/RCode stick, | >the 32MB AiboMind stick will not work) | | No, I am not using a different memory stick. From the manual it wasn't clear to | me I had to have a different memory stick for the ERS-7. Where is the OS for the | robot coming from if not using the Aibo Mind stick? Shouldn't it be able to | coexist with the object I am going to write? | | If I need an extra memory stick, where do people usually get? Sony? Then if the | robot rebots using my own memory stick all the functionality of Aibo minds is | missing. Isn't there a way to have from it some of its functionality intermixed | with my own app? | | More specifically. I want to build a server on my linux box that will build a | image based 3D model of my house from the images Aibo is capturing. From there I | want Aibo to expand its awareness of its environment. So for example, if he is | in the kitchen and I tell him "goto your room" he can get directions from the | Linux server how to get there and view the camera get course corrections. | | >See the installation guide where it talks about 'telnet' | > | >The built-in Windows telnet will do (or HyperTerminal will also work). Connect | >to port 59000 for the 'console' | >For the normal IP address: | > telnet 10.0.1.100 59000 | > | I get it! thanks. | | >>I try | >>finding something on the AIBO LAN Manager (included on the bot CD) | > | >That's for the AiboMind software, I don't recommend you use it. | > | >[it can be used to help configure the WLANCONF.TXT, but IMHO if you are | >programming at the OpenR SDK level, you should be able to edit the | >WLANCONF.TXT file yourself] | | That is correct. However the LAN manager came handy to double check how it save | the wep key into the WLANCONF.TXT file. | |
| +- 0870 RE:HOWTO LAN Console on ERS-7 AiboPet [2004/01/03-03:10] |
| | Response : Edit : Delete |
|
| >>I
assume you are using a different memory stick for this (a PMS/RCode
stick, | >>the 32MB AiboMind stick will not work) | > | >No, I am not using a different memory stick. | | [a common incorrect assumption - as you point out it is not clearly explained] | | You need what is called a "Programmable Memory Stick" (or PMS or RCODE stick) | It is pink and does *not* have an ERF- number on it. | | You can buy them from SonyStyle or RoboToys.com | $39 for the ERA-MS016 [16MB stick] | | http://www.aibohack.com/rcode/index.html [read the "Please Read This"] | | ------- | >Where is the OS for the | >robot coming from if not using the Aibo Mind stick? | | The version of the OS used for the OpenR SDK is a sub-set of the AiboMind | functionality. The files you copy when you setup the PMS stick provide the | core OS and basic OpenR functionality. It takes up only a small portion of the | PMS stick (ie. an 8MB or 16MB PMS stick has lots of room for your own | program/data) | | > Shouldn't it (ie. AiboMind) be able to | >coexist with the object I am going to write? | | Sorry, that's not possible. | It has been often requested, but still not possible for technical and | marketing reasons (ie. Sony's decision). The bulk of the AiboMind software is | encrypted. Even the Aibo Custom Manager add-in logic is encrypted. | | See the graphic at the bottom of this page for an explanation of the different | worlds. In this graphic, "AiboMind" ~= "AiboLife" and "SDK app" means the | OpenR SDK. | http://www.aibohack.com/openr_sdk/index.html | Many features of AiboMind are not available to OpenR SDK programs. | | > Then if the | >robot rebots using my own memory stick all the functionality of Aibo minds is | >missing. | | Correct. | | > Isn't there a way to have from it some of its functionality intermixed | >with my own app? | | The things not included: walking, voice recognition, card recognition, email, | ball tracking, ... | You can re-write them from scratch, but it can be a lot of work. | Right now there is no built-in way of making the ERS-7 walk using the OpenR | SDK, so just getting the ERS-7 to walk well is a lot of work. | | ========== | >More specifically. I want to build a server on my linux box that will build a | >image based 3D model of my house from the images Aibo is capturing. From | there I | >want Aibo to expand its awareness of its environment. So for example, if he is | >in the kitchen and I tell him "goto your room" he can get directions from the | >Linux server how to get there and view the camera get course corrections. | | If you want to capture the video, you can use the AiboMind Plus extension. | However, that is a passive process. You can't influence AIBO's logic, or add | new voice commands like "goto your room". | | For the general solution, you may be better off waiting an announcement of an | RCODE or RCODE-like or RCodePlus programming tool [ie. these exist for the | older model ERS-210[A] etc, but not for the ERS-7, yet]. | If you want to get started now, I'd recommend buying a ERS-210[A]. It is 3+ | years old and has more programming options. The ERS-7 platform is brand new. | |
| +- 0871 HOWTO LAN Console on ERS-7 angel [2004/01/04-00:38] |
| Response : HomePage : Email : Edit : Delete |
|
Terrific
information. Thanks for your help. I already have an ERS-7 so I will try to program as much as I can. Since the OpenR gives me access to sound and video capture I am not worried about speech and facial recognition. I can do that code and there are plenty of libraries out there for it. The most time consuming step is going to program the motion. I wasn't counting on having to do that this early. Now I see why Sony provides a Lighwave model and the joints specs, it will come handy. Thanks |
| V 0866 Multiple UDP connections gkhealy [2003/12/31-13:18] |
| Response : Edit : Delete |
|
I
am writing a UDP server which should be able to respond to multiple client connections concurrently The sample UDPEchoServer program looks like it should support this however EndPoints after the first give the error: UDPEchoServer::DoStart() : Bind() fail[n] If UDPECHOSERVER_CONNECTION_MAX is made greater than 1 I have traced this to an error from the UDPEndPointBind message which returns error code 4 which the manual indicates is IP_CONNECTION_CLOSED. Any help would be appreciated. thanks, Gerry |
| V 0855 New documentation? bara [2003/12/17-08:45] |
| | Response : Email : Edit : Delete |
|
| Are
there any release dates for the 1.1.4 documentation? We've been having
some | problems with the new format used by the camera. | | Thanks, | Alex | |
| |- 0857 RE:New documentation? openrsupport@(Administrator) [2003/12/17-12:07] |
| | | Response : Edit : Delete |
|
| | Alex, | | | | The 1.1.4 documentation will be released soon, but there will only be minor | | changes. It will update a file name and fix a minor error in the install guide. | | | | If you want to know the new format (FBK=Filter Bank) used by the camera, please | | refer to the sample programs that use the camera image. | | | | You can find "Filter Bank" in a signal processing textbook or a web search. | | | | Thank you, | | | | OPEN-R SDK Support | | |
| | +- 0864 RE:New documentation? bara [2003/12/19-01:44] |
| | Response : Edit : Delete |
|
| >Alex, | > | >The 1.1.4 documentation will be released soon, but there will only be minor | >changes. It will update a file name and fix a minor error in the install guide. | > | >If you want to know the new format (FBK=Filter Bank) used by the camera, please | >refer to the sample programs that use the camera image. | > | >You can find "Filter Bank" in a signal processing textbook or a web search. | > | >Thank you, | > | >OPEN-R SDK Support | | I have been able to solve my problem. It was that somebody changed the openr | system files in the Memeory Stick with the previous version - 1.1.3 - while I | was compiling with 1.1.4 (Doh!). And thank you for the pointer, right now I am | checking the filter bank reference. | | Thanks, | Alex | |
| +- 0859 RE:New documentation? AiboPet [2003/12/18-01:22] |
| | Response : Edit : Delete |
|
| >... We've been having some | >problems with the new format used by the camera.... | | What's the problem ? | | The provided sample code show how to merge the different bank layers into a | monochrome image. I have code that does it for color. | | FWIW: AiboSpeedRacer uses a generic module (VisionU) that captures the 6 banks | of the hires camera [ERS-7 or ERS-2x0] converts it to RGB, compress it to | JPEG, and finally transmits it to a PC using UDP/IP. | http://www.aibohack.com/openr_sdk/speedracer.htm | | The VisionU module can be used separately. | |
| +- 0863 RE:New documentation? bara [2003/12/19-01:41] |
| Response : Edit : Delete |
|
>>...
We've been having some >>problems with the new format used by the camera.... > >What's the problem ? > >The provided sample code show how to merge the different bank layers into a >monochrome image. I have code that does it for color. > >FWIW: AiboSpeedRacer uses a generic module (VisionU) that captures the 6 banks >of the hires camera [ERS-7 or ERS-2x0] converts it to RGB, compress it to >JPEG, and finally transmits it to a PC using UDP/IP. >http://www.aibohack.com/openr_sdk/speedracer.htm > >The VisionU module can be used separately. Thank you, for the application I'm doing that will probably be enough. =) Alex |
| V 0861 wireless switch shutdown (ers-7) ejt [2003/12/18-14:52] |
| Response : HomePage : Email : Edit : Delete |
|
just
curious if anyone knows how to prevent the automatic shutdown on an
ERS-7 if you flip the wireless switch on the underside...? Might be nice to be able to use that switch for something other than another power button ;) |
| V 0856 ERS-7 Model Information Updated openrsupport@(Administrator) [2003/12/17-10:05] |
| Response : Edit : Delete |
|
Hello, Model Information for ERS-7 has been updated from the Dec. 9 version. https://openr.aibo.com/openr/eng/perm/download/download.html Changes from 20031209-E-001-B2 to 20031215-E-001-B3 Page3 lower right figure: front leg distance from Tilt3 center to paw ground point Change: 68.5 To: 71.5 lower right figure: rear leg distance from Tilt3 center to paw ground point Change: 77.26 To: 76.5 Page4 the upper side The following are added. Figure: Measurements of External Appearance of front/rear legs Thank you, OPEN-R SDK Support (The article that 2003/12/17-10:02 posted, was modified by Author) |
| V 0851 ERS-7 speaker volume ejt [2003/12/15-13:50] |
| | Response : HomePage : Email : Edit : Delete |
|
| It
seems that even though I set the volume of the speaker to 'ospkvol10dB', | which is the largest documented volume level, it's still nowhere near as loud as | when I run the Aibo Mind software. Is there a trick to it that I'm missing? | I'm following the example of the SoundPlay sample (which I updated for the | ERS-7, but it too is rather soft.) | | thanks, | -ethan | |
| +- 0852 RE:ERS-7 speaker volume ejt [2003/12/15-14:23] |
| | Response : HomePage : Email : Edit : Delete |
|
| A
little more checking around answered my own question: | | From OPEN_R_SDK/OPEN_R/include/OPENR/OPrimitiveControl.h: | ospkvolinfdB = 0x8000; // Volume 0 (min) | ospkvol25dB = 0xe700; // Volume 1 | ospkvol18dB = 0xee00; // Volume 2 | ospkvol10dB = 0xf600; // Volume 3 (max) | | So, the value being passed as the volume setting is actually the direct dicibel | gain, not just an enumeration. The values being passed are encoded as a two's | complement short representing the amplification in decibels. | | So, passing 0 will give you full volume. A positive value would theoretically | give you amplification, but experimentally, this doesn't really seem to make it | any louder... looks like 0 will make things as loud as they're going to get. | | But if anyone wants to play with fading sounds in and out, you could do it by | incrementing the decibel gain with finer control than the constants would imply | is possible. | | (caveat: on my 210, sounds get a little clipped when using full volume, so it | would only really be useful if you're going to be demoing in a noisy room. on | the ers-7 however, it still sounds good. In fact, I'd like to point out the | ERS-7's speakers sound significantly better in general, and i'm not even an | audiophile! ;) | |
| +- 0853 RE:ERS-7 speaker volume openrsupport@(Administrator) [2003/12/16-08:43] |
| Response : Edit : Delete |
|
Hello, We strongly recommend that you not set volumes larger than (ospkvol10dB = 0xf600;). 10dB is the maximum safe volume for AIBO's circuitry. You can set a volume greater than 10dB, but it might cause damage to AIBO's circuitry; when AIBO plays such a loud sound for an extended period, increased current causes a higher temperature in AIBO's body, which can lead to damage. Please keep in mind that the OPEN-R SDK has minimal restrictions regarding AIBO's hardware, so it is recommended to follow this advice. Thank you, OPEN-R SDK Support |
| V 0850 TINYFTPD cannot copy DESIGNDB,CFG, CONNECT,CFG files nswamy [2003/12/10-10:50] |
| Response : HomePage : Email : Edit : Delete |
|
Hello, I am trying to incorporate TinyFTPD into all the sample programs by changing the OBJECT.CFG with an additional line that includes TINYFTPD.BIN. I did this thinking that it may be easy for me to use TINYFTPD and download different trees into the memory stick using a wireless network. The sample HELLOWORLD has only one config file "OBJECT.CFG" and so I can download that tree into the Memory stick anytime. But there are some sample programs that need a "DESIGN.CFG", a DESIGNDB.CFG etc. When ever I have one of these programs running on the AIBO, subsequent TinyFTP downloads fail if the new tree also has the same file(s) - design.cfg, DesigbDB.cfg. Is it possible that DESIGNDB.CFG and CONNECT.CFG in OPEN-R/MW/CONF directories get locked and it is not possible to have them overwritten? The error I see is shown below... $ ~/mstreeput 192.168.1.224 MS ./OPEN-R/ ./OPEN-R/MW/ ./OPEN-R/MW/CONF/ ./OPEN-R/MW/DATA/ ./OPEN-R/MW/OBJS/ ./OPEN-R/MW/CONF/CONNECT.CFG ./OPEN-R/MW/CONF/DESIGNDB.CFG 553 error. Any pointers appreciated. Thanks, nanu |
| V 0838 Just saying Hey.... Adam4444 [2003/12/02-16:13] |
| | Response : Edit : Delete |
|
| Are
all of you in here computer programmers? Seems like there are about 5
of us | with Aibo's. I am pretty tech savy...but hardly a programmer...after I download | the required software and such....will it be crazy to learn? I have | patience...and a good dog. = ) | | Thanks for any response.....looking forward to the little beast learning some | new tricks. | | Adam | |
| +- 0839 RE:Just saying Hey.... TodboT [2003/12/03-09:03] |
| | Response : Edit : Delete |
|
| Hi Adam, | | I'm more of an engineer than a programmer. I just like to toy with the robot, | but nothing major software-wise. I did chop off its legs and put big wheels on | to see if it could be done. If you don't have much programming experience(C++), | then getting started with the SDK will be pretty difficult. I'd suggest | starting with some of the higher level tools (R-CODE or AiboPet's YART). | | R-CODE: http://www.us.aibo.com/openr_code.htm | YART: http://www.aibohack.com/rcode/yart.htm | | Tod | |
| +- 0840 RE:Just saying Hey.... sapient [2003/12/04-20:55] |
| | Response : HomePage : Email : Edit : Delete |
|
| Tod, | I'm intrigued by your comment : | | >> I did chop off its legs and put big wheels on | | And was wondering how you did it? is there any info out there about this kind of | thing?(googling doesn't help me) Does this mean that you understand the protocol of the | connections? | | Andrew | |
| +- 0841 RE:Just saying Hey.... TodboT [2003/12/05-02:54] |
| | Response : Edit : Delete |
|
| Hi
Andrew, | | I don't know the protocol. It was just a hack. First you have to remove the | leg from the body. If you know how to turn a hobby servo into a continuous | rotation motor, then it's the same idea. The limiter, which prevents Aibo's leg | from rotating all the way around, needs to be removed. This is the only *real* | permanent modification. Then the potentiometer is centered and left free so it | doesn't rotate. Now, when you try to move the leg, it will continually spin. | The farther you try to move the leg from center, the faster it will rotate. The | only other trick is to fake a connection to the missing potentiometers in the leg. | | Here's a video I took a while ago: | http://www.aibo-life.com/visorg/video/BigWheelAibo.mpg | | If you're interested in doing this, let me know and I can try to help you out. | | Tod | |
| +- 0844 RE:Just saying Hey.... openrsupport@(Administrator) [2003/12/05-10:46] |
| | Response : Edit : Delete |
|
| Hello, | | Please be aware that the modifications mentioned in this thread | are not officially sanctioned by Sony. Feel free to discuss these | issues, but keep in mind that these posts should not be interpreted | as official advice to other AIBO owners. | | Best regards, | | OPEN-R SDK Support | |
| +- 0845 Monster truck dog sapient [2003/12/09-03:27] |
| | Response : HomePage : Email : Edit : Delete |
|
| Thanks
go to admin for letting me continue this thread. | | This kind of mod is definitely questionable in a number of ways. And I personally would | advise everyone *NOT* to mod thier aibo - bad things lie that way. | | Still... | | Like the first monkeys they shot into space, I think that my aibo may too become a hero | (martyr?) sooner or later. Does that make me a bad person? | | I'd also like to point out that I think it's a great pity that ERS7 is no longer modular. But, | since it may represent the end of Sony profitting from the Aibo 'inter-module protocol' | perhaps it might be a chance to have the protocol released?! | | I know a few hifi enthusiasts, they wouldn't dream of buying all of thier equipment from | one company, a common protocol ties their modules together. Perhaps this would help | the emerging industry, allowing a more diverse set of "aibo" standard products; | benefitting Sony in the long run. | | Gosh, I sound like a regular fanatic! Anyone else agree/disagree? | | Andrew | |
| +- 0849 RE:Monster truck dog AiboPet [2003/12/10-02:45] |
| Response : Edit : Delete |
|
>
Like the first monkeys they shot into space, I think that my aibo may
too become a hero >(martyr?) sooner or later. Does that make me a bad person? I prefer ICybies... (much cheaper) http://www.aibohack.com/icybie/spares.jpg > I'd also like to point out that I think it's a great pity that ERS7 is no longer modular. The underlying modular CPC architecture is still there. Same for the ERS-31x dogs, and even the original ERS-11x dogs. Now (like the ERS-31x) you have to take the dog apart to get access to the modules. >since it may represent the end of Sony profitting from the Aibo 'inter-module protocol' Who knows whether the ERS-220 upgrade kit was profitable ;-> I bought one, but the whole idea was not well exploited IMHO. Externally swappable modules can cause more connection problems than they are worth -- so overall I think they made the right decision with the ERS7 |
| V 0847 Communication between dogs Malleus12 [2003/12/09-13:34] |
| | Response : Email : Edit : Delete |
|
| Hello
I'm a student working on a thesis with aibos and we are currently trying | to get multiple dogs to talk to eachother over the wireless LAN. We are | pretty early on in the project and could use a little guidance in this area. | Specifically right now we want to get one dog to tell another to move a | certain distance in a specific direction (like in the telnet program where you | can tell a dog to walk x distance with y rotation). Is there a way we can | just quickly modify the EchoServer/EchoClient code to allow this type of | communication or is there something more complicated involved? Any help is | much appreciated, thanks. | |
| +- 0848 RE:Communication between dogs AiboPet [2003/12/10-02:36] |
| Response : Edit : Delete |
|
... >Specifically right now we want to get one dog to tell another to move a >certain distance in a specific direction ... Is there a way we can >just quickly modify the EchoServer/EchoClient code to allow this type of >communication or is there something more complicated involved? Nothing more complicated - that's the way to do it. The EchoServer/Client samples have all the structure in place. You just need to define the data messages sent between the robots (in place of the echoed string). I assume you already have code to do the actual walking. You will need to have one AIBO act as the 'server' and the other dogs act as clients. The IP address for the server AIBO will have to be hard coded in each of the clients. That's the easiest setup IMHO, but there are other variations possible (ie. use a PC as a centralized server, have each dog be both a client and server...) Another important thing is to be sure you have an Access Point in the room (on the same radio channel/net). As each AIBO is booting up, it will check to see if there is a WiFi signal in the room. Ad-hoc AIBO to AIBO without an access point has a slim chance of booting/connecting. |
| V 0846 ERS-7 Model Information openrsupport@(Administrator) [2003/12/09-09:24] |
| Response : Edit : Delete |
|
Hello, Model Information for ERS-7 has just been posted. You can find it in the download area. https://openr.aibo.com/openr/eng/perm/download/download.html Thank you, OPEN-R SDK Support |
| V 0806 Problems about MoNet bigjingo [2003/11/11-16:12] |
| | Response : Email : Edit : Delete |
|
| Recently,
I am learning OPENR by reading the tutorial and corresponding code | from the sample provided. When I was using the MoNet section something went | wrong. I added the commands defined in the file /sample/MoNet/MS/OPEN-R/MW/ | CONF/MONETCMD.CFG into /sample/MoNet/MoNetTest/MoNetTest.h, and modified the | original function MoNetTest::ReadyCommand() in file /sample/MoNet/MoNetTest/ | MoNetTest.cc. After booting, however, Aibo does not act as I designed. It | seems that the problem is that the function ReadyCommand() excutes only once, | then no ASSERT_READY message return to the function, and thus no message | circulation occurs. | Is there anybody who have such experience or know what to do? Thanks a lot. | |
| +- 0843 RE:Problems about MoNet openrsupport@(Administrator) [2003/12/05-07:02] |
| Response : Edit : Delete |
|
>
ReadyCommand() excutes only once Yes, that's right. void MoNetTest::ReadyCommand(const OReadyEvent& event) { if (moNetTestState == MNTS_START) { MoNetCommand cmd(SLEEP2SLEEP_NULL); subject[sbjCommand]->SetData(&cmd, sizeof(cmd)); subject[sbjCommand]->NotifyObservers(); moNetTestState = MNTS_WAITING_RESULT; } } MoNetTest::ReadyCommand works only in the START section. This function puts AIBO into the SLEEP posture, and does nothing after that. If you want to have your code circulated, you should put it in a suitable function. Which function you choose varies depending on what you want to do. Best regards, OPEN-R SDK Support |
| V 0814 How to make the team markers? zhangzy [2003/11/17-12:35] |
| | Response : Email : Edit : Delete |
|
|
Problem just as the title. Please hint me which material is used to make | the team markers and how to change the uniform easily. Which color should be | use? | |
| +- 0842 RE:How to make the team markers? openrsupport@(Administrator) [2003/12/05-06:59] |
| Response : Edit : Delete |
|
>
Problem just as the title. Please hint me which material is used to make >the team markers and how to change the uniform easily. Which color should be >use? Hi, are you asking about team markers for RoboCup usage? If so, please consult the RoboCup website or the Four-Legged mailing list. You can see this page for more information. https://openr.aibo.com/openr/eng/perm/university.html Best regards, OPEN-R SDK Support |
| V 0835 "abort() called"? curufin [2003/11/29-04:58] |
| | Response : Edit : Delete |
|
| When
I'm running my program after a while it says "abort() called" in the
telnet | terminal and then there's no more output to the terminal. Anyone know what | causes this? | |
| +- 0837 RE: abort called from where AiboPet [2003/11/30-01:58] |
| Response : Edit : Delete |
|
>...
Anyone know what causes this? Smart-a** answer: Probably someone calling the 'abort()' routine ;-> When debugging shutdown, startup or hanging problems AIBO may crash before all the necessary 'printf' and other data have been sent via telnet. Several suggestions to hunt this down. Try providing your own 'abort()' routine to replace the runtime version. Have it printf a warning in a loop so you will be sure to see it on the telnet display. [Also removing abort from the library will show you other places where it may be called] If you really need low level debugging, you can solder to the inside of your AIBO to export the serial port signals (gives you the same output as the telnet session, but immediately even if the LAN is not working) [BTW: I have done this to my AIBOs, so if you can't get any further, and you have a self contained and reproducable case, please send me an email and I can try it out to see if there is more info printed] |
| V 0831 How to set the camera resolution raco60 [2003/11/27-21:35] |
| | Response : Email : Edit : Delete |
|
| I am
modifing the ImageCapture OPENR example, and I don't know how and who | especifies the values of OFbkImageInfo->width and height. Also, I don't find | which is the maximum resolution of the camera. Thank you very much | |
| +- 0832 RE:How to set the camera resolution AiboPet [2003/11/28-01:56] |
| | Response : Edit : Delete |
|
| >I
am modifing the ImageCapture OPENR example, and I don't know how and who | >especifies the values of OFbkImageInfo->width and height. | | You don't set the resolution. The camera ALWAYS captures all possible image | resolutions as multiple 'layers'. It is up to you to extract which layer you | want to use for processing: | | H Layer - 176x144 (or 208x160 for the ERS-7) | M Layer - half that 88x76 (or 104x80 for the ERS-7) | L Layer - half that again 44x38 (or 52x40 for the ERS-7) | | The C Layer (color detection) is the same resolution as the M layer. One byte | per pixel. | | --- | With the latest SDKs (1.1.4 for 2x0 and 1.1.5 for ERS-7) it captures | additional bands where you can combine to build a double-H layer (called "HR" | layer in the sample) | | Synthetic "HR" Layer - 352x244 (or 416x320 for the ERS-7) | | ------ | > Also, I don't find | >which is the maximum resolution of the camera. | | The HR Layer is also the native (and maximum) resolution of the camera | 352x244 for the ERS-2x0 | 416x320 for the ERS-7 | |
| |- 0833 RE:How to set the camera resolution AiboPet [2003/11/28-01:59] |
| | Response : Edit : Delete |
|
| [hard
to edit on this BBS] | There are some math/typo errors on the last post (eg: 2*144 should equal 288). | You hopefully get the point. Everything is in ratios of 1:2 or 1:4 or 1:8 of | the native camera resolution | | ERS-2x0: 352 x 288 | ERS-7: 416 x 320 | |
| +- 0834 RE:How to set the camera resolution ejt [2003/11/28-14:38] |
| | Response : HomePage : Email : Edit : Delete |
|
| >The
HR Layer is also the native (and maximum) resolution of the camera | >352x244 for the ERS-2x0 | >416x320 for the ERS-7 | Actually, I've been meaning to ask about this. I've noticed in several technical spec sheets, such as | this one: | http://www.aibo-europe.com/1_1_3_ers7_specifications.asp?language=en | that the camera resolution is noted at 350,000 pixels for the ERS-7. Last I checked, | 416*320=133,120 | However, 720*480=345,600 | So actually, I was expecting the resolution to be 720x480 when I first got the specs. | | So, I'm guessing the native camera resolution is much higher than what's accessible in software. | There could be many reasons for that - that's a lot of pixels to push around at 30Hz, but it would | be nice to have access to it for a snapshot or other intermittant processing. *shrug* Just curious. | |
| +- 0836 RE:How to set the camera resolution AiboPet [2003/11/30-01:50] |
| Response : Edit : Delete |
|
FWIW:
According to CAMERA.INF, the ERS-7 camera is the Fujitsu MB86S03 (set to 416x320 pixels). Some websites say the MB86S03 camera can do 640x480 and other resolutions (not sure about 720x480). The camera is only one part of the problem. They have specialized hardware to do the multi-resolution filter bank and color range detection. Not to mention all the software algorithms... |
| V 0825 Floating-point exception Nasnah [2003/11/21-19:07] |
| | Response : Edit : Delete |
|
| I
get a floating point exception in a numeric calculation (using double).
Could the | reason be a number to close to zero, and if so how could I avoid that? | | The double in $f00-$f01 looks denormalized to me (but I'm not that good on hex)... | Is there any reference documentation on double on AIBO? | | Excerpts from emon.log: | exception code: 15 Floating-Point exception | | $f00: 0x00000000 $f01: 0x00000019 | $f20: 0x21eca015 $f21: 0xbfdfad83 | | Assembler instruction that causes the crash: | 411628: 4620a002 mul.d $f0,$f20,$f0 | |
| +- 0828 RE:Floating-point exception AiboPet [2003/11/23-02:34] |
| Response : Edit : Delete |
|
>I
get a floating point exception in a numeric calculation (using double). Could the >reason be a number to close to zero, and if so how could I avoid that? >The double in $f00-$f01 looks denormalized to me (but I'm not that good on hex)... >$f00: 0x00000000 $f01: 0x00000019 I agree, this looks highly suspicous [ie. looks more like integers loaded from memory instead of a floating point double] >Is there any reference documentation on double on AIBO? AIBO uses the standard MIPS instruction set (MIPS III for floating point) There are any number of MIPS instruction set references that can be found on the web (including mips.com ) For floating point, the double precision registers contain the upper part in $f01 and the lower part in $f00 $f01:$f00 = 1 sign bit:11 exponent bits:52 fraction bits [64 bits total] If you can find the source code that generates the MUL.D instruction, print the values before the calculation -- I suspect you will see a very small number (less than 1E-308 which is not properly normalized) |
| V 0817 Sending large amounts of data? RCRegion? curufin [2003/11/19-22:53] |
| | Response : Edit : Delete |
|
| I
would like to send a large array of integers between objects, which is
the best way | to do this? Apparently (or so I've read), messages have a max size which probably | is smaller than what would be needed for my array. | | I've started to look into RCRegion and how to use this for passing objects via | shared memory. However, I can't seem to find a good tutorial on how to do this for | arbitrary data (i.e. _not_ for sending commands to joints etc.) If I would like to pass | my large int-array (or a custom (c++) object) this way, how should my code look? | | Code examples or pointers to useful tutorials would be greatly appreciated. | |
| |- 0818 RE:Sending large amounts of data? RCRegion? ejt [2003/11/20-06:34] |
| | Response : HomePage : Email : Edit : Delete |
|
| >
[OPEN-R] messages have a max size | I'm not sure about this off hand, it would be good to know what it is.... Anyone? | | > I would like to send a large array of integers between objects, which is the | > best way [to send] my large int-array (or a custom (c++) object) | int-arrays are much simpler, and just an extension of what you'll find in the | OPEN-R sample code. However, you mention sending C++ objects, which can be | rather... interesting. So I'll go into that, and along the way, you can extract | the couple lines of code you'd need if you just want to send an int array. | | The first question you need to ask yourself is, do you want to send a single | type of class, or do you want to allow polymorphism? (i.e. you're sending Shape | objects, but the actually classes sent are subclasses like Rectangle or Ellipse, | and further a Circle is a subclass of Ellipse...) | | The first case, if you're sending a single, static type, or even different types | but you're going to use some manual form of transmiting type information as | well, then the process isn't much different than sending an array of ints - you | can just put the object into the RCRegion's memory buffer, and on the receiving | end you'll be able to typecast it back and call the object's functions. The key | here is you're NOT using virtual functions. You tell the compiler to interpret | the data as a particular type, and so it will assume that's the type it actually | is and everything should be happy. | | However polymorphic classes imply virtual functions. The problem is, you're | just telling the compiler that, for instance, it's a Shape, and so when you call | area() on it, how does it know whether to run Rectangle::area() or | Ellipse::area()? There's some extra information hidden by the compiler (Run | Time Type Information - RTTI) that lets it figure out what the objects *actual* | type is, and call the appropriate function. | | The problem with this is that the receiving and sending OObject are compiled | independently (no fork() system call :(), and so that RTTI information differs. | between them. The sender marks it as a Shape, but the receiver will read the | RTTI and try to interpret it as who-knows-what. | | There is a workaround - you must compile the sender and receiver code together, | and then duplicate the binary and run two copies of it, one as sender, and one | as receiver. This is a little hackish, but it works. You can check out the | implementation details in the MMCombo -> MainObj/MotoObj set up in Tekkotsu. | This lets us use generic MotionCommand objects which can be subclassed as Walk | or HeadPointer or MotionSequence, etc... | | Now, general pointers: | Be careful of byte alignment. If you copy data into a buffer, and then try to | read a non-even word aligned double, or non-word aligned ints or floats, it'll | crash. In general, make sure you start copying on an even word boundary | (address divisble by 8). | If you want to put an object into a buffer, don't initialize the object and then | copy it - set it up directly *in the buffer*. | So instead of: | ....Obj foo(...); //or Obj * foo = new Obj(...), same idea | ....// Setup foo | ....memcpy(buf,&foo,sizeof(foo)); | Do this: | ....Obj * foo = new (buf) Obj; //foo==buf | ....// Setup foo | That second method uses the valid, but lesser known variant of new which will | initialize an object at a pre-allocated memory address, which can be handy! | | And finally some links: | | SharedObject<class T> - encapsulates a class of T in an RCRegion, rounded to | page size | http://www-2.cs.cmu.edu/~tekkotsu/dox/SharedObject_8h-source.html | | SharedQueue<unsigned int maxsize, unsigned int maxentries> - another way to move | data between OObjects... we use this to send events between OObjects (e.g. Sound | play complete, MotionCommand finished) | http://www-2.cs.cmu.edu/~tekkotsu/dox/SharedQueue_8h-source.html | | Of course, depending on what you're doing, the easiest way might be to let our | framework do it for you ;) | http://www.tekkotsu.org/ | |
| +- 0821 RE:Sending large amounts of data? RCRegion? AiboPet [2003/11/21-00:59] |
| | Response : Edit : Delete |
|
| >I
would like to send a large array of integers between objects, which is
the | best way | >to do this?... | | The *easiest* (and slimy-est) way to do this is to use the 'nomemprot' memory | option (where there is no memory protection between objects/tasks) and just | past pointers to data willy-nilly. | | The *best* way to do this is to use the C++ class 'RCRegion' (for simple data | structures that don't have function pointers or virtual functions as ejt | points out) | | Look at the RCRegion.h header for general use of the class. | For different examples of RCRegion, just look at the standard examples for | allocation, access to data (rgn->Base(), rgn->Size() etc), and how to send to | other objects (SetData(rgn)) | | If you don't like the C++ class, you can use the more primitive shared memory | functions [OPENR::NewSharedMemoryRegion] | |
| +- 0823 RE:Sending large amounts of data? RCRegion? ejt [2003/11/21-05:21] |
| | Response : HomePage : Email : Edit : Delete |
|
| >The
*easiest* (and slimy-est) way to do this is to use the 'nomemprot'
memory | >option (where there is no memory protection between objects/tasks) and just | >past pointers to data willy-nilly. | Yeah, I should have pointed this out... I'm used to trying to stay mostly | compatable with memprot mode, even though our software runs in nonmemprot mode | (it's faster) | | But, don't forget you can still run into the RTTI problems I mentioned earlier | if one OObject accesses an object in another OObject's memory space. This is a | general problem of accessing objects created in a different process (err, | OObject), regardless of how/where it's currently stored. | | So I hate to recommend against using object oriented code, because it can make | life so much easier for some things, but it adds a lot of complication for this | stuff. | | Of course, if someone could add a fork() command and/or multithreading commands | to the API, then this problem would go away... (hint, hint ;) | |
| +- 0826 RE:Sending large amounts of data? RCRegion? AiboPet [2003/11/22-01:42] |
| Response : Edit : Delete |
|
>Of
course, if someone could add a fork() command and/or multithreading commands >to the API, then this problem would go away... (hint, hint ;) I wouldn't hold my breath on that. --- If you can keep everything to short routines (ie. and not rely on preemptive object task switching -- the normal restriction within the one .BIN file) then just linking all your objects into the same .BIN file makes the most sense. This also reduces the number of extra copies of runtime functions and any shared memory hassles. |
| V 0822 AiboSpeedRacer (ERS-2x0, ERS-7) AiboPet [2003/11/21-01:59] |
| Response : Edit : Delete |
|
SpeedRacer
will to turn your AIBO into a remote control toy. You can see what AIBO is seeing using the full resolution of the camera. http://www.aibohack.com/openr_sdk/speedracer.htm Please read the warnings. This uses the CMU knee walk and has more wear-and- tear on your AIBO's joints. ===== Technical comments: Built using the OpenR SDK The differences between the two versions (ERS-2x0 and ERS-7) are only in the base version of the OpenR system binaries. Comprises three modules: BENCH.BIN - walkbench module which takes telnet single character commands to make AIBO walk MOTION.BIN - CMU CMPack02 MotionObject trimmed down to use the Knee Walk VISIONU.BIN - UDP server that handles the telemetry requests, including full size camera images compressed to JPEG, camera settings etc. NOTE: The older WalkBench supports both the CMU knee walk and the Sony OMWares walk styles (similar to the ERS-11x and AiboLife1). The Sony OMWares modules do not work on the ERS-7 (yet) Email me if you want a snapshot of the current source. I still want to add more features (including CDT color images, information like distance sensors, battery levels and temperatures, ...) |
| V 0815 ERS-7 camera resolutions dvail2 [2003/11/18-08:38] |
| | Response : Email : Edit : Delete |
|
| Hi, | | We're looking at the resolution of the camera in the ERS-7. Using the old | interface, the highest resolution that we're able to access is 208x160. Is a | higher resolution mode available? | | thanks, | -Doug | |
| +- 0816 RE:ERS-7 camera resolutions AiboPet [2003/11/19-02:04] |
| Response : Edit : Delete |
|
>We're
looking at the resolution of the camera in the ERS-7. Using the old >interface, the highest resolution that we're able to access is 208x160. Is a >higher resolution mode available? Yes, the camera is twice that 416x320 (4x as many pixels) To get access to the higher resolution data, you need to use 6 different bands and combine them to make a 416x320 image. Normally you only need 3 different bands (Y, Cr, Cb). The 3 extra bands combined with the Y band will give you 4 times as many pixels. Source code for the monochrome version that merges the bands is in the ERS-7 SDK ImageObserver sample. See 'ReconstructImage' I have a color version that also does JPG compression and sends the image data to a PC program for display (416x320 around 5 fps). Coming soon in 'AiboSpeedRacer'. It also uses the CMU 'knee walk' http://aibohack.com/pix/speedracer_teaser.jpg |
| V 0809 "Downloader::Download cannot open module file"? Nasnah [2003/11/14-19:48] |
| | Response : Edit : Delete |
|
| When
I try to boot AIBO the following error message is written to the telnet
console and AIBO | shuts down. Both Map and Navigator are written by me, and OPEN-R parts of the code (DoInit, | DoStart, etc.) are identical. Any ideas as to what may be causing this? Maybe errors in the | config files? | | [oid:0x80000048] /MS/OPEN-R/MW/OBJS/MAP.BIN | Downloader::Download cannot open module file : /MS/OPEN-R/MW/OBJS/NAVIGATOR.BIN. | [oid:80000031,prio:1] OObjectRegistry::LoadObjects() : | Load(/MS/OPEN-R/MW/OBJS/NAVIGATOR.BIN) error 90 | OPowerManager::PowerOff() SHUTDOWN ... | |
| +- 0810 RE: AiboPet [2003/11/15-00:56] |
| | Response : Edit : Delete |
|
| >
Both Map and Navigator are written by me, and OPEN-R parts of the code | (DoInit, | >DoStart, etc.) are identical. Any ideas as to what may be causing this? Maybe | errors in the | >config files? | | My guess: | Check both the .ocf files. If they have the same object name, that's your | problem. Each .ocf file must have a unique name. | |
| +- 0812 RE: Nasnah [2003/11/15-22:27] |
| | Response : Edit : Delete |
|
| >>
Both Map and Navigator are written by me, and OPEN-R parts of the code | >(DoInit, | >>DoStart, etc.) are identical. Any ideas as to what may be causing this? Maybe | >errors in the | >>config files? | > | >My guess: | >Check both the .ocf files. If they have the same object name, that's your | >problem. Each .ocf file must have a unique name. | | Thankyou for your help, but the problem seems to be something else. | The Navigator.ocf looks like: | object Navigator 3072 16386 128 cache tlb user | and Map.ocf looks like: | object Map 3072 16386 128 cache tlb user | |
| +- 0813 RE: AiboPet [2003/11/16-02:05] |
| Response : Edit : Delete |
|
Oops,
should have picked this up the first time "NAVIGATOR.BIN" is too long. Filenames should be limited to 8.3 names [eg: rename it to "NAVIGATR.BIN" or something like that] |
| V 0808 need help programming ERS-7 "game mode" hounshell [2003/11/13-09:52] |
| | Response : HomePage : Email : Edit : Delete |
|
| I'd
like to know how to program one of the free "game-mode" slots for the
ERS- | 7 (using the included "aibo mind" stick). | | I'm an AI researcher and interested in "hooking-up" the ERS-7 to a | distributed network of computers. Our company has developed some | software for making distributed publish/subscribe and directory very | easy using C++. We want to integrate this network "grid-processing" | environment with the ERS-7's existing behavior base, if possible using | one of the "game mode" slots for "aibo mind". We intend to test our | pattern processing platform, (providing a "history" for the robot). | | Already our ERS-7 has the beta Open-R SDK installed. It works fine. | but I saw no mention how to interface new programs with the | existing aibo mind software. But I'm new to Open-R, and may have missed it. | | any help would be greatly appreciated. | | thanks. | |
| +- 0811 RE:need help programming ERS-7 AiboPet [2003/11/15-01:01] |
| Response : Edit : Delete |
|
>I'd
like to know how to program one of the free "game-mode" slots for the
ERS- >7 (using the included "aibo mind" stick). I'll let the official openrsupport people give you the reasons why, but the short answer is "you can't" --- A little background: There has always been a gap between what I call "commercial AiboWare" (including AiboLife, AiboMind, Navigator and even the RCODE interpreter) and the free and open OpenR SDK (for non-commercial use) The goals of the OpenR SDK are different, and as such you can't mix-and-match the two (see recent question about adding OpenR objects to Navigator, which can't be done either) http://www.aibohack.com/openr_sdk/aiboware.gif (in the first column replace "AiboLife2" with "AiboMind" - and you'll get the point) |
| V 0807 ERS-7 3D model data has been released openrsupport@(Administrator) [2003/11/13-02:30] |
| Response : Edit : Delete |
|
Hi, We have just released 3D model data for ERS-7 under the 'Downloads' area. For those of you interested in using a 3D representation of AIBO in your software, you can find ERS-210, ERS-220, and now ERS-7 model data in our downloads area. Enjoy! OPEN-R SDK Support |
| V 0803 EchoServer and EchoClient dividedsky319 [2003/11/11-06:02] |
| | Response : Email : Edit : Delete |
|
| I'm
trying to get 2 dogs to talk to each other, I'm trying to figure out how | exactly it's done. How might I go about doing this by using EchoServer and | EchoClient? I don't understand exactly what they do, and where I would specify | the IP address to speak to, etc. I'm still a beginner at all this, it took a | while to get the network configuration correct for me to even contact them via | the wireless network but finally got that step figured out. Any tips? | Thanks. | |
| +- 0804 RE:EchoServer and EchoClient dividedsky319 [2003/11/11-06:33] |
| | Response : Email : Edit : Delete |
|
| >I'm
trying to get 2 dogs to talk to each other, I'm trying to figure out how | >exactly it's done. How might I go about doing this by using EchoServer and | >EchoClient? I don't understand exactly what they do, and where I would specify | >the IP address to speak to, etc. I'm still a beginner at all this, it took a | >while to get the network configuration correct for me to even contact them via | >the wireless network but finally got that step figured out. Any tips? | >Thanks. | | I found that there's a place to put the address to my EchoServer in | EchoClientConfig.h , yet even when I change it to the correct address, when I | telnet to the Client to see how the bootup is going and what's going on in the | dog, it says "EchoServer_IP is 10.0.1.2 (EchoClientConfig.h)" | | This is after I delete all the bin files in the directory, run make and make | install once I edit EchoClientConfig.h. Any ideas? Thanks. | |
| +- 0805 RE:EchoServer and EchoClient TodboT [2003/11/11-07:08] |
| Response : Edit : Delete |
|
Try
doing a make clean. |
| V 0798 OPEN-R SDK for ERS-7 (beta) has been released openrsupport@(Administrator) [2003/11/05-02:50] |
| | Response : Edit : Delete |
|
| OPEN-R
SDK for ERS-7 (1.1.5-beta1) and samples have been released. | You can download them from the Download area. | | Thank you, | | OPEN-R SDK Support | |
| +- 0802 RE:OPEN-R SDK for ERS-7 (beta) has been released AiboPet [2003/11/11-02:58] |
| Response : Edit : Delete |
|
>OPEN-R
SDK for ERS-7 (1.1.5-beta1) and samples have been released. Thanks for the early release. It works great. A friendly reminder, the Soccer Lion sample needs to be ported/adapted. It relies on the OMWares modules that are not available in source form. Thanks in advance |
| V 0800 Problems using SDK with ERS-220A BerserkLehigh [2003/11/08-03:43] |
| | Response : Email : Edit : Delete |
|
| I
have been attempting to implement open-r SDK with an AIBO 220A, but
have been | having immense difficulties. After compiling sample programs and copying the | appropriate files to a memory stick, when the AIBO is turned on the green power | light comes on, stays on for 15 or so seconds, and then goes out. With | programs prepared through Sony software the AIBO reacts as expected. | | I suspect that the AIBO does not recognize the format in the memory stick, or | that the system files are not correct using open-r SDK, but I am unclear on how | to resolve this issue. Thank you for any help you can offer. | |
| +- 0801 RE:Problems using SDK with ERS-220A ejt [2003/11/08-06:19] |
| Response : HomePage : Email : Edit : Delete |
|
If
you are using the WCONSOLE system files, you can telnet into the aibo
on port 59000 before it is finished booting and get some idea where it was when it failed. You can also check the /ms/open-r/emon.log file after the crash for more information on the cause of the crash. After you press the power button to start booting, if you start repeatedly trying to telnet in, at first the connection will be refused, then it will make a connection but no output will appear (kill the connection and try again), and eventually, the boot will proceed far enough you'll actually connect and get text. Just keep trying to connect while it's booting up so you can get connected ASAP. Keep in mind however, that the console output may drop a few lines that were still in the networking buffers when the crash occured, so they don't get displayed. So if you're trying to track down the location of a crash, you'll need to put each debugging line inside a loop so it will be displayed a hundred times or so to force the buffer to clear before moving to the next section of code. Also, there are a few utilities available for parsing the emon.log file to generate a stack trace. We include the tool from the Mi-Pal robosoccer team ( http://gucis.cit.gu.edu.au/~mi-pal/ ) although other teams have implementations as well. (The article that 2003/11/08-06:16 posted, was modified by Author) |
| V 0795 Using the AIBO's speaker ESP [2003/11/01-14:03] |
| | Response : Email : Edit : Delete |
|
| Hi All, | | I am streaming audio back to my ERS-210 with UDP. In the examples, it shows how | to load an play a song from the memory stick. Instead each time a chunck of | data (byte array) is received from the server, I want my AIBO to play it from | it's speaker. Format of audio etc... sent from server is all correct, and sent | at a rate of 2048 bytes/32ms (I can change these to whatever I need). Any | ideas on the speaker side of things? Thanks guys. | |
| +- 0799 RE:Using the AIBO's speaker TodboT [2003/11/07-03:18] |
| Response : Edit : Delete |
|
You
should be able to write the incoming audio into a buffer, and when the
speaker needs to play the next chunk of audio (when Ready() or CopyWAVTo() is called in the SoundPlay example), copy the data from the buffer into the RCRegion used to play sound. If you want to playback audio at 16KHz 16bit MONO, then you'll want to send 1024 bytes/32ms. Keep us updated! |
| V 0796 Aibo over LAN not working dividedsky319 [2003/11/03-09:12] |
| | Response : Email : Edit : Delete |
|
| I
went through the HelloWorld instructions in the installation guide, and
did | a "make" and "make install" ... I copied the OPEN-R directory from | WCONSOLE/MEMPROT/ to the memory stick, and then copied the OPEN-R directory | from the MS directory for the HelloWorld program onto the memory stick, | overwriting the objects.cfg. I then edited the WLANCONF file in the | SYSTEM/CONF/ directory with the proper information (Using the same exact | WLANCONF information someone used last year in the same room for a thesis) | Yet, when I boot up the dog, I can't even ping the dog. Did I do everything | correctly? Do you think it's in the network configuration files that something | is wrong? Perhaps something was changed... (However, the static IP for aibo1, | aibo2, etc. is still the same) Maybe my gateway changed? Would this mess | things up? Thanks. | |
| +- 0797 RE:Aibo over LAN not working AiboPet [2003/11/04-00:58] |
| Response : Edit : Delete |
|
>I
went through the HelloWorld instructions in the installation guide, and
did >a "make" and "make install" ... I copied the OPEN-R directory from >WCONSOLE/MEMPROT/ to the memory stick, and then copied the OPEN-R directory >from the MS directory for the HelloWorld program onto the memory stick, >overwriting the objects.cfg. I then edited the WLANCONF file in the >SYSTEM/CONF/ directory with the proper information (Using the same exact >WLANCONF information someone used last year in the same room for a thesis) >Yet, when I boot up the dog, I can't even ping the dog. Did I do everything >correctly? Do you think it's in the network configuration files that something >is wrong? Perhaps something was changed... (However, the static IP for aibo1, >aibo2, etc. is still the same) Maybe my gateway changed? Would this mess >things up? Thanks. You have listed many of the possible problems. WiFi setup can be a pain. Also the WCONSOLE version is designed to be connected with telnet before the actual program starts running. The "WLAN" version will also work (for the minimal 'ping' test) Have you tried the original (unedited) WLANCONF.TXT file ? (10.0.1.100, AIBONET, WEP enabled "AIBO2") ? Can you post the WLANCONF.TXT file you are using ? What are the settings on the PC WiFi card / access point ? |
| V 0794 Author DELETE ***** [2003/11/01-14:04] |
| Response : Edit : Delete |
|
(The
article that 2003/11/01-13:55 posted, was modified by Author) |
| V 0792 How can I get the "Y_Cr_Cb table"?? sheepman [2003/10/28-17:28] |
| | Response : Email : Edit : Delete |
|
| Hello. | | The sample program "BallTrackingHead" offers the color table for "PINK BALL." | | The color table contains "Y-Cr_max-Cr_min-Cb_max-Cb_min" format. | | for example, | | cdt->Set( 0, 230, 150, 190, 120); | ... | cdt->Set(31, 230, 150, 190, 120); | | I found the other table for "blue color" | | for example, | | cdt->Set( 0, 131, 51, 229, 149); | ... | cdt->Set(31, 131, 51, 229, 149); | | I know the fact that AIBO can distinguish 8 colors simutaneously. | | So I want to use 4 colors, but I don't know another table. | | Can you recommand for this problem? | | How can I get the another Y-Cr-Cb table? | | Thank you for read this. | |
| +- 0793 RE:How can I get the AiboPet [2003/10/29-00:21] |
| Response : Edit : Delete |
|
>Hello. >... >I know the fact that AIBO can distinguish 8 colors simutaneously. > >So I want to use 4 colors, but I don't know another table. One easy way it to take the color ranges from the existing AiboWare. It is in the TR.CT file (not encrypted). It gives more hand-tuned color ranges that appear to work better for some of the commercial AiboWare. For example the very old AiboLife1 stick has several built-in colors: Red, Pink, blank, Flesh, Blue, Greeny. Many newer sticks have fewer colors (focusing on Red + Pink and Flesh) http://www.aibohack.com/210/colors.htm http://www.aibohack.com/210/color_w01.jpg >How can I get the another Y-Cr-Cb table? The TR.CT file format is relatively simple, and I wrote a program to dump it and adapted it to Open-R C++ format. See code snippet here: http://www.aibohack.com/a/tr_ct_dump_out.txt |
| V 0790 Where can i find the open-r for the ERS-7 slider2nl [2003/10/26-07:16] |
| | Response : Email : Edit : Delete |
|
| Where can i find the open-r for the ERS-7 | | i have search it but can't find it ???? | |
| +- 0791 RE:Where can i find the open-r for the ERS-7 AiboPet [2003/10/28-01:15] |
| Response : Edit : Delete |
|
>Where
can i find the open-r for the ERS-7 > It hasn't been released yet. The SDK is supposed to be available around the time the robots are released. The robots are expected "Delivery around 11/21" in the USA according to http://www.us.aibo.com |
| V 0788 New to programming Aibos - wireless help? dividedsky319 [2003/10/25-03:05] |
| | Response : Email : Edit : Delete |
|
| Myself
and 2 other undergrads are doing our thesis on programming Sony Aibos. | Someone did a thesis on the same topic last year, and somewhat got one dog to | work correctly (localize itself, go towards the goal, kick the ball). But what | we hope to do is get a team playing, requiring a transfer of information | between the dogs wirelessly. Does anyone know of anyplace I could look for | information on how to get started? Any help would be appreciated. Thanks. | |
| +- 0789 RE:New to programming Aibos - wireless help? AiboPet [2003/10/26-02:15] |
| Response : Edit : Delete |
|
For
AIBO-to-AIBO communication, using TCP/IP over WiFi is relatively easy. The EchoServer/EchoClient samples how you can set this up. There is also a UDP version of the sample (TCP/UDP). As far as higher level world modelling, and the data you trade between the AIBOs, that is a much more difficult problem [specific to your thesis I assume]. I'd suggest looking at the RoboCup entries with public source, some have AIBO-to-AIBO communication (I know some use sound. Unsure about WiFi use in the current offerings) |
| V 0786 New Image layers? ejt [2003/10/17-08:57] |
| | Response : HomePage : Email : Edit : Delete |
|
| >FBK(Filter Bank) Full Band support. | >- New luminous images (Y_LH, Y_HL, and Y_HH) that | Is there someplace with more information about these images? I'm not sure what | the LH, HL,and HH stand for... (and thus I'm not sure how to interpret the data) | |
| +- 0787 RE:New Image layers? AiboPet [2003/10/18-00:43] |
| Response : Edit : Delete |
|
>>FBK(Filter
Bank) Full Band support. >>- New luminous images (Y_LH, Y_HL, and Y_HH) that >Is there someplace with more information about these images? I'm not sure what >the LH, HL,and HH stand for... (and thus I'm not sure how to interpret the data) My suggestion: Look at the code for ImageObserver::ReconstructImage() in the new ImageObserver sample It shows how you expand 4 image pixels from each input pixel from the 4 layers. It does B&W only. I haven't tried merging in the color layers yet (UV/CrCb) |
| V 0785 SoccerLion200-r2.tar.gz has been released openrsupport@(Administrator) [2003/10/17-05:53] |
| Response : Edit : Delete |
|
Dear OPEN-R SDK users, SoccerLion200-r2.tar.gz has been released. It works with OPEN-R SDK 1.1.4-r1. You can download it in the "Download" section. Thank you, OPEN-R SDK Support |
| V 0783 Loading a *.bin file in Tekkotsu Version 1.5 cyberclone3 [2003/10/14-20:51] |
| | Response : Email : Edit : Delete |
|
| Hello, | | How Can I load a *.bin file in tekkotsu v1.5? Is it possible? I have several | *.bin objects files that work together in order to AIBO follows a line. | | Thanks, | | ToNi. | |
| +- 0784 RE:Loading a *.bin file in Tekkotsu Version 1.5 ejt [2003/10/15-00:00] |
| Response : HomePage : Email : Edit : Delete |
|
>How
Can I load a *.bin file in tekkotsu v1.5? Is it possible? I have several >*.bin objects files that work together in order to AIBO follows a line. It should be possible - they can be put directly on the memory stick like the system .bin's. Put the .bin's in the <project>/ms/open-r/mw/objs/ directory and you'll need to modify <project>/ms/open-r/mw/conf/*.cfg as usual to set up the connections and load the .bins you want. Also, if you wanted to integrate the sources for those into a Tekkotsu project, it's a little more tricky, but the Makefile has some comments about what would need to be changed to have it compile the new OObjects (aka processes aka .bin's) alongside the current ones. |
| V 0766 Open-R on Wiindows 98 / Linux tarx [2003/10/07-23:27] |
| | Response : Email : Edit : Delete |
|
| Hi everybody, | | I am planning to set up a Aibo-Class at my local High-School (in Germany). All | we have ist Windows98 and/or Linux. The Installation-Manual says that OpenR | needs XP or Win2000. Any way to make it work on 98 or is Linux the better choice? | | Greetings | |
| +- 0769 RE:Open-R on Wiindows 98 / Linux ejt [2003/10/08-02:05] |
| | Response : HomePage : Email : Edit : Delete |
|
| I
would personally tend to recommend linux... You'll just wind up working
in cygwin on windows | anyway, might as well get the "real thing" if it's an option. But really, it's pretty close - I'd say go | with whichever you're more comfortable with and whichever has better reliability in a lab | environment with a bunch of high schoolers. | | I'm not sure if Win98 can work or not... it all depends on whether cygwin will run in it, since all the | compilation is done using Makefiles and cygwin. | | You might want to look at our Tekkotsu project as well - I used it in a summer program for High | School Juniors this past summer, for some of them it was their first time programming... (not that I | would recommend it as such - that was a little rough for the newbies to pick up in the 5 weeks... | but if you have more time, you could teach a pretty thorough course on Object Oriented | programming using it.) | http://www.tekkotsu.org/ | (The article that 2003/10/08-02:01 posted, was modified by Author) | |
| +- 0780 RE:Open-R on Wiindows 98 / Linux tarx [2003/10/13-17:33] |
| | Response : Edit : Delete |
|
| >I
would personally tend to recommend linux... You'll just wind up working
in | cygwin on windows | >anyway, might as well get the "real thing" if it's an option. But really, it's | pretty close - I'd say go | >with whichever you're more comfortable with and whichever has better | reliability in a lab | >environment with a bunch of high schoolers. | > | >I'm not sure if Win98 can work or not... it all depends on whether cygwin will | run in it, since all the | >compilation is done using Makefiles and cygwin. | > | >You might want to look at our Tekkotsu project as well - I used it in a summer | program for High | >School Juniors this past summer, for some of them it was their first time | programming... (not that I | >would recommend it as such - that was a little rough for the newbies to pick | up in the 5 weeks... | >but if you have more time, you could teach a pretty thorough course on Object | Oriented | >programming using it.) | >http://www.tekkotsu.org/ | >(The article that 2003/10/08-02:01 posted, was modified by Author) | | Thank you. As I as I experienced it now, the cygwin environment does not comply | to Win98, but the first steps went well with Linux (Suse 8.2). Your project | looks great and I would like to use it with my class. As my enlgish isn't the | best, I am not sure wether the Matlab-Software is required or not. Do i | understand right that the code is Matlab independent as long as I have the Java | installed? Does your framework require a certain version of Linux? Thanks in | advanced and sorry to bother you again. | |
| +- 0782 RE:Open-R on Wiindows 98 / Linux ejt [2003/10/14-14:43] |
| Response : HomePage : Email : Edit : Delete |
|
>I
am not sure wether the Matlab-Software is required or not. Do i Matlab is not required. In fact, the Matlab tools are actually currently broken, but there's an easy work around. If anyone really cares, they probably know enough to fix it properly and send us a patch. ;) Java isn't required either for that matter - control is also possible from the console or just the Aibo's own buttons. But it's nice to use the GUI tools. >Does your framework require a certain version of Linux? Anything that will compile the SDK itself should be fine, which is pretty much everything in the last few years probably. We've been running Red Hat 7, 9, Mac OS X, and a recent SuSE... |
| V 0781 Problems with Spray client cyberclone3 [2003/10/13-20:34] |
| Response : Email : Edit : Delete |
|
Hello again, I have problems when I do a make from cygwin in order to create the *.class file. I get many errors. In cygwin environment, from the sprayClient-29 directory I do a make and get 81 errors. I already have installed Quicktime for java and java 1.4.1_01 SDK. Thanks, ToNi. |
| V 0778 Flash updater problem... akent [2003/10/13-04:49] |
| | Response : Edit : Delete |
|
| I
have a 210 (non-A) that does not accept the 1.1.3-r2 firmware update. | It plays a sad tune about 7 seconds after power on, and then it turns off. | | I know it needs the update, as Open-R SDK generated programs lock up, and I | must use a pin to then remove the battery. | | I have tried this with two different memory sticks (ERA-MS016), and used | the .exe as well as the gzipped updater versions, both sticks cleaned of all | files (not formatted). I extracted the updater files to the root of the memory | stick. The stick contains only the root directory 'upgrade', along with files | and directories under that. | | I also used two different fully charged batteries. | | All other programs (Aibopet's gems, Dog's Life, and the factory sticks work | fine. | | Any ideas on what I can try? I sure would like to use the Open-R SDK. | | Thanks, | Kent Farnsworth | |
| +- 0779 RE:Flash updater problem... akent [2003/10/13-07:44] |
| Response : Edit : Delete |
|
Teach
me to read the directions!!! Sorry to bother y'all. The problem is fixed. You can't just install the updater directly onto the stick! After doing that with Aibopet's wonderful stuff, I assumed the world worked that way... I feel like an idiot... Take care, Kent |
| V 0777 Tekkotsu Release **1.5** ejt [2003/10/11-06:25] |
| Response : HomePage : Email : Edit : Delete |
|
We
are pleased to announce the 1.5 release of the Tekkotsu Development
Framework. http://www.tekkotsu.org/ This release contains many new features and bug fixes, and relatively few changes in the APIs. Support for the new OPEN-R SDK release is also included (backwards compatable with previous release as well) A summary of the changes made is given below. For the full version history, please visit http://www.tekkotsu.org/VersionHistory.html # New Features: ... * Segmenting Threshold training tool for vision ......... o http://www.tekkotsu.org/CameraSetup.html ... * Significant improvement of noise detection during segmentation ... * Watchable memory ......... o http://www.tekkotsu.org/TekkotsuMon.html#wmVar ......... o Provides ability to display current values and receive ............. events when they are changed ... * Now includes Mi-Pal crash analysis tools from Griffith Univ. ......... o http://gucis.cit-student.gu.edu.au/%7Emi-pal/ ......... o Provides stack trace after a crash using the emon.log file. ... * Build system reorganized ......... o Distributed builds available ... * Contribution from Pennsylvania Governor's School for the ....... Sciences, 2003 Computer Science team project: "See, Spot; Run" ......... o Runtime WalkMC parameter editor (imported into mainline) ......... o Line following (not imported into mainline) ......... o New pace and crawl gaits: ............. http://www.tekkotsu.org/Samples.html#walking ... * Controller GUI scripts implemented (formerly "bookmarks") ......... o Any script(s) named "STARTUP" will be run the first time ............. the GUI connects ......... o Any script(s) named "CONNECT" will be run every time the ............. GUI makes a connection ... * Added speed limiter to HeadPointerMC to help protect neck ....... joints. Also added config variables to control maximum speed. ....... See max_head_*_speed in project/ms/config/tekkotsu.cfg ... * Colorized errors/warnings in build output (run tools/cmake ...... instead of make) ......... o default colors are best for reverse video (white text on ............. black) but you can change them easily in tools/colorfilt ... * You can now type option names as well as indexes on the console # API Changes: ... * Reorganized TekkotsuMon using Java packages ......... o Better Aibo 3D integration ......... o Now uses java.util.prefs to remember window positions ......... o breaks MATLAB tools - see known bugs ... * Remove need for vision->enableEvents() calls ... * Aibo3D Monitor and Aibo3D Controller have been merged # Bug Fixes: ... * Support for the new version of OPEN-R SDK 1.1.4-r1 ......... o only minor change was needed regarding introduction of new ............. image channels ... * Controls no longer crash if they have no submenus but return ....... non-NULL upon activation ... * Finally killed occasional ControllerGUI's ArrayIndexOutOfBounds ... * Behaviors based on StateNode no longer hang on launch ... * Outgoing connections now can receive data (thanks to Andrew ....... Cristina and the group from UNO.) ... * If a header file is missing, the build process will give a nice ....... error and stop instead of repeatedly making the first dependency ....... file that refers to it. ... * Fixed possible crash if shutting down while playing sound ... * Blocking output (serr) now gives correct output - had been ...... repeating output, issue with non-blocking remains (see known ...... bugs) ... * Memory is now freed properly after saving an image sequence from ...... VisionGUI # Known Bugs: (you can help!) ... * Blocking output doesn't actually block - it just isn't buffered ....... as much? This might be a system limitation? ... * Unretained StateNodes are unstable - doesn't do reference counting ....... properly somewhere ... * MATLAB links to Java components is broken - Java stuff is in ....... packages now, MATLAB can't find appropriate files. ......... o To solve, does anyone know how to direct MATLAB to find a ............. class in a package? ......... o To workaround, may simply need to move Listener.class ............. files back into same directory as MATLAB's .m files ............. (tools/mon). ... * When a new WalkMC is started, it has a tendancy to try to ....... straighten the legs first ... * In Aibo3D, the knee joint position shown is offset from the real ... * If you stop and restart the WMMonitorBehavior, the GUI will lose ....... all the wmVar's. |
| V 0776 Streaming Audio tiabo [2003/10/10-18:08] |
| Response : Email : Edit : Delete |
|
Hi, I have an ERS-210, and am working on some software to stream audio from the AIBO's microphones to a server, have the server convert to a playable format, and return to the AIBO, to be output via it's speaker. I have managed to do all of this using UDP to send audio to the server, but I have run into some issues sending back to AIBO. I cannot work out how to stream audio out of the speaker. Instead I have only been able to get it to load an existing file from the MS. This means sending a file back from the server via TCP, saving to MS and then loading to speaker. Ultimately I would like to send back to the AIBO using UDP and stream straight to the speaker. The sample apps have proven useful in the past, but I see nothing about streaming audio. Any suggestions where to look or how to do it? Cheers. |
| V 0773 Problems With Spray cyberclone3 [2003/10/09-19:12] |
| | Response : Email : Edit : Delete |
|
| Hello, | | I have a problem with the application Spray when compiling and linking with | cygwin. I do the following: | | From cygwin, in the directory networkedImageObserver I do a make and it | executes Makefile file. The error is: | | Toni@SISINTELI00 /cygdrive/c/spray/NetworkedImageObserver | $ make | /usr/local/OPEN_R_SDK/bin/mipsel-linux-g++ -O2 -g -I. - | I/usr/local/OPEN_R_SDK/OP | EN_R/include/R4000 -I/usr/local/OPEN_R_SDK/OPEN_R/include/MCOOP - | I/usr/local/OPE | N_R_SDK/OPEN_R/include -o NetworkedImageObserver.o -c NetworkedImageObserver.cc | NetworkedImageObserver.cc: In member function `OStatus | NetworkedImageObserver::DoStartSpray()': | NetworkedImageObserver.cc:106: parse error before `=' token | make: *** [NetworkedImageObserver.o] Error 1 | | What is the problem? Is the proces I do incorrect? What's wrong? | | Thanks, | | ToNi. | |
| +- 0775 RE:Problems With Spray AiboPet [2003/10/10-03:13] |
| Response : Edit : Delete |
|
>Hello, > >I have a problem with the application Spray when compiling and linking with >cygwin. There is a bug in the header. Look for the definition of "DEFAULT_PORT" and change it to be just a number Note: this is a relatively old sample, so you you may have other build problems. |
| V 0774 Author DELETE ***** [2003/10/10-02:57] |
| Response : Edit : Delete |
|
(The
article that 2003/10/10-02:20 posted, was modified by Author) |
| V 0772 Cygwin and OPEN-R SDK 1.1.4-r1 openrsupport@(Administrator) [2003/10/09-01:41] |
| Response : Edit : Delete |
|
Dear all, We released Cygwin 1.3.22 on this website, which can be downloaded from the "Download" section. To use the latest OPEN-R SDK (1.1.4-r1) on Windows, you need this version of Cygwin installed. Previous versions of Cygwin are not compatible with the Remote Processing functionality in the latest OPEN-R SDK. "gdb" is included in this version of Cygwin. You can now debug OPEN-R objects using Remote Processing; please see Tutorial 7 on the OPEN-R SDK University page for more information. Thank you, OPEN-R SDK Support |
| V 0770 OPEN-R SDK 1.1.4-r1 has been released openrsupport@(Administrator) [2003/10/08-05:38] |
| | Response : Edit : Delete |
|
| * OPEN-R SDK 1.1.4-r1 has been released. | You can get it from the download page. | | FBK(Filter Bank) Full Band support. | - New luminous images (Y_LH, Y_HL, and Y_HH) that | are separated by band filters can be retrieved from | the image sensor in addition to the Y, Cr, and Cb images. | You can reconstruct a double resolution luminous image | from the Y, Y_LH, Y_HL, and Y_HH images. | | RP (Remote Processing) OPEN-R | - TCPGW on the host program executes much faster than before. | - RP OPEN-R patch for GCC3.2 on Linux is included. | | IPv4 protocol stack | - Support DHCP | - WEPKEY can now be set in hexadecimal. | | Environmental variable OPENRSDK_ROOT can be set | | * OPEN-R SDK sample 1.1.4-r1 has been released. | | ImageObserver and W3AIBO can now reconstruct image data | with double the resolution than previous samples. | | Also, added a new sample program [BattChecker] that shows | the remaining charge level of AIBO's battery. There are | three ways that you can use BattChecker: with AIBO only, | as a simple Win32 program, or as a Win32 program with | shared memory and MFC. | | * jpegsrc.v6b.tar.gz can be downloaded. It is used in the | W3AIBO sample program. | | Thank you, | | OPEN-R SDK Support | |
| +- 0771 RE:OPEN-R SDK 1.1.4-r1 has been released AiboPet [2003/10/09-00:53] |
| Response : Edit : Delete |
|
> >FBK(Filter Bank) Full Band support. Cool! |
| V 0765 Another answer to AiboPet cyberclone3 [2003/10/07-18:52] |
| | Response : Email : Edit : Delete |
|
| Hello again, | | Is it very difficult to implement it (control remote camera) in my own program | made with OPEN-R? Does it require too much time? Can you tell me where Can I | find information about how to do it? | | Thanks, | | ToNi. | |
| +- 0768 RE:Another answer to AiboPet ejt [2003/10/08-01:51] |
| Response : HomePage : Email : Edit : Delete |
|
>Is
it very difficult to implement it (control remote camera) in my own
program >made with OPEN-R? Does it require too much time? Can you tell me where Can I >find information about how to do it? If you build on top of Tekkotsu (which itself builds on top of OPEN-R) then many features, such as remote controlling the camera, and a vision monitor) are already built in and available... http://www.tekkotsu.org/ |
| V 0764 Answer to Dzitter about Spray and Answer to AiboPet.Thanks cyberclone3 [2003/10/07-18:37] |
| | Response : Email : Edit : Delete |
|
| Hello, | | Is it Spray made by Dzitter freeware? And, AiboPet, is it your freeware? | Where Can I find information about how do it? | | Thanks. | |
| +- 0767 RE:Answer to Dzitter about Spray and Answer to AiboPet.Thanks AiboPet [2003/10/08-00:55] |
| Response : Edit : Delete |
|
>Hello, > >Is it Spray made by Dzitter freeware? Dzitter's website is here: http://www.zitter.net/aibo/ The spray / "Network ImageObserver" class is what you want. http://www.zitter.net/aibo/spray/index.html The sample code is for an older version I believe (88x72 25fps, no compression) With using my YUV10 codec, you can get even faster larger images (and 176x144) > And, AiboPet, is it your freeware? >Where Can I find information about how do it? RCodePlus is on my site, it uses the YUV10 codec. http://www.aibohack.com/rcode/index.html HOWEVER, it is not an OpenR SDK program. It uses the YUV10 video codec. I have the source to it. The source for decoding is in the LAN samples. |
| V 0761 Loading program made with OPEN-R into user motion menu ..... cyberclone3 [2003/10/06-19:07] |
| | Response : Email : Edit : Delete |
|
| Hello, | | I have a program made with OPEN-R, and I would like to load it into user | motion menu of AIBO Navigator in order to see what the AIBO is seeing with the | camera when he does some tasks, for example, when he follows a line. How can I | do it? | | Thanks. | |
| +- 0762 RE:Loading program made with OPEN-R into user motion menu ..... AiboPet [2003/10/07-01:24] |
| | Response : Edit : Delete |
|
| Sorry, you can't. See: | | https://openr.aibo.com/cgi-bin/openr/e_regi/im_trbbs.cgi? | uid=general&df=bbs.dat&prm=TAN&pg=1&no=0759#0759 | | However you can incorporate Navigator-like features into your own OpenR | program. For remote camera, this means creating your own video codec and PC | viewer program. I have similar ones for RcodePlus (using TCP/IP). Dzitter | wrote one a similar one called 'spray' (using UDP/IP). | You can do similar things if you want to listen in on what AIBO is hearing too. | |
| +- 0763 RE:Loading program made with OPEN-R into user motion menu ..... AiboPet [2003/10/07-01:28] |
| Response : Edit : Delete |
|
URL
garbled. See reply 0759 below. |
| V 0760 Univ. of Pennsylvania Robocup code ddlee [2003/10/03-13:13] |
| Response : HomePage : Email : Edit : Delete |
|
The
Univ. of Pennsylvania Robocup source code, including OPEN-R modules, embedded Perl interpreter, and Perl scripts are now available from our website (http://www.cis.upenn.edu/robocup/) from the following link: http://www.cis.upenn.edu/robocup/source/UPenn03.tar.gz It is being released under the GPL license. A technical description regarding these routines will be available shortly. Best regards, -Dan |
| V 0757 AXF File Format cyberclone3 [2003/09/26-19:36] |
| | Response : Email : Edit : Delete |
|
| Hello,
I have a little problem. How can I convert a *.bin executable file to a | *.axf file format in order to load it in the user motion menu of AIBO | Navigator? | | Thanks. | |
| +- 0759 RE:AXF File Format AiboPet [2003/09/27-00:51] |
| Response : Edit : Delete |
|
>Hello,
I have a little problem. How can I convert a *.bin executable file to a >*.axf file format in order to load it in the user motion menu of AIBO >Navigator? > [and similar second question] You can't. They are two different things. .BIN files are program code, that can do potentially anything. .AXF file are like .ODA file (I believe the format is the same too). They contain just motion data (packed .MTN files) that can be played back in certain situations when triggered by Navigator. ---- Do you use .MTN files inside of your program for motion playback ? See the ODA file format reader in the MoNet sample if interested in going that route. |
| V 0758 AXF File Format cyberclone3 [2003/09/26-19:40] |
| Response : Email : Edit : Delete |
|
How
Can I generate a *.axf file format from my source code .cc with OPEN-R?
Is it possible? Thanks |
| V 0756 ERS-7 and ERS-210 in RoboCup 2004 openrsupport@(Administrator) [2003/09/25-02:05] |
| Response : Edit : Delete |
|
Dear all, We have posted a new FAQ. Q. Can I join RoboCup 2004 by using ERS-7? Can I join RoboCup 2004 by using ERS-210? A. Rules for the RoboCup Four-Legged League are decided by the Four-Legged League technical committee. The Technical Committees of the league officially decided 2004 competition platforms for the league as below. - Entertainment Robot AIBO ERS-210(A)/B - Entertainment Robot AIBO ERS-7 For more detail, please visit "Sony Four-legged Robot League official site for RoboCup2003" for more information: http://www.openr.org/robocup/index.html Thank you, OPEN-R SDK Support |
| V 0752 Is there anything wrong with the robot? ddnt [2003/09/20-01:33] |
| | Response : Edit : Delete |
|
| All problems we developed got the emon.log as following: | | mCOOP exception | [exception info] | time stamp: 0x00000000f7fb2c88 | status:$12: 0x0000ff13, cause:$13: 0x80000008 | badvaddr: $8: 0x216b280d, epc:$14: 0x20a27268 | [object info] | context: 0x80210240, state: 0x00000001 | last context: 0x00000000,last thread: 0x00000000 | [initial value info] | sr:$12: 0x0000ff13 | [register dump] | at: r1: 0x20a356c8, v0: r2: 0x216b0410, v1: r3: 0x00002124, a0: r4: 0x0000000c | a1: r5: 0x00000037, a2: r6: 0x00000000, a3: r7: 0x216b280d, t0: r8: 0x00000024 | t1: r9: 0x00000000, t2:r10: 0x0000ff13, t3:r11: 0x00000004, t4:r12: 0x000000ff | t5:r13: 0x00002100, t6:r14: 0x00000008, t7:r15: 0x00002108, s0:r16: 0x00000000 | s1:r17: 0x00000000, s2:r18: 0x216b280d, s3:r19: 0x215b0000, s4:r20: 0x00000037 | s5:r21: 0x00000024, s6:r22: 0x20d0fadc, s7:r23: 0x20a00000, t8:r24: 0x00000001 | t9:r25: 0x00000037, gp:r28: 0x20d08000, sp:r29: 0x209d2e88, s8:r30: 0x00000000 | ra:r31: 0x20a27270, hi: : 0x00000000, lo: : 0x000001cc | | exception code: 2 TLB exception (load or instruction fetch) | | [system control co-processor registers] | Index 0x00000005 Random 0x00000010 EntryLo0 0x0001a79e EntryLo1 0x00000000 | Context 0x0010b590 PageMask 0x00000000 Wired 0x00000002 ??? ---------- | BadVAddr 0x216b280d Count 0x4364baa1 EntryHi 0x00000000 Compare 0xf806724a | Status 0x2000ff00 Cause 0x10009c2c EPC 0x800ba824 PRId 0x00002831 | Config 0x204766f3 LLAddr 0x3881436d ??? ---------- ??? ---------- | XContext 0x0010b590 ??? ---------- ??? ---------- ??? ---------- | ??? ---------- ??? ---------- ECC 0x00000044 CacheErr 0xa4200000 | TagLo 0x00000000 TagHi 0x00000000 ErrorEPC 0xbfc00c50 ??? ---------- | [floating-point general registers] | $f00: 0x00000000 $f01: 0x00000000 $f02: 0x00000000 $f03: 0x3fe00000 | $f04: 0x00000000 $f05: 0x3ff00000 $f06: 0x00000000 $f07: 0x3fe00000 | $f08: 0x00000000 $f09: 0x40634000 $f10: 0x00000000 $f11: 0x4079b000 | $f12: 0x606a63be $f13: 0x3f7a98ef $f14: 0x00000000 $f15: 0x417c9c38 | $f16: 0x00000000 $f17: 0x4079b000 $f18: 0x00000000 $f19: 0xc18c9c38 | $f20: 0x01942359 $f21: 0x1714079e $f22: 0x388912ab $f23: 0x13cb534e | $f24: 0x00f52bdc $f25: 0x50922b6e $f26: 0x0a149b97 $f27: 0x370c7406 | $f28: 0x6f76187e $f29: 0x479b10a5 $f30: 0x81ce9f93 $f31: 0x176128a9 | [floating-point control registers] | $0:fp_eir: 0x00002830 | $31:fp_csr: 0x01000000 (bit[17-0]: ----------------RN) | [object list] | Name ContextID ThreadID ExecSpcID OID MetaSpace | ------------------------------------------------------------------------------- | systemCore 0x80202c20 0x80202dc0 0x80202d80 ---------- mCore | systemCore(1) 0x80203a20 0x80203b80 0x80202d80 ---------- mCore | mCOOPReflector 0x80207200 0x80208a00 0x80208200 0x8000000b mCore | uniMailer 0x80207360 0x80208a60 0x80208240 0x8000000c mCore | mCOOPFaultHandler 0x802074c0 0x80208ac0 0x80208280 0x8000000d mCore | registryManager 0x80207620 0x80208b20 0x802082c0 0x8000000e mCore | addressManager 0x80207780 0x80208b80 0x80208300 0x8000000f mCore | kernelModeLib 0x802078e0 0x80208be0 0x80208340 0x80000010 mCore | mCoreReflector 0x80207a40 0x80208c40 0x80208380 0x80000011 mCore | idle 0x80207ba0 0x80208ca0 0x802083c0 0x80000012 mCore | exceptionHandler 0x80207d00 0x80208d00 0x80208400 0x80000013 mCore | analyzer 0x80207e60 0x80208d60 0x80208440 0x80000014 mCore | mDriveReflector 0x80207fc0 0x80208dc0 0x80208480 0x80000015 mCore | obletManager 0x8020cfc0 0x80208e20 0x802084c0 0x80000016 mCore | installer 0x8020ce60 0x80208e80 0x80208500 0x80000017 mCore | driverManager 0x8020cd00 0x80208ee0 0x80208540 0x80000018 mCore | objectManager 0x8020cba0 0x80208f40 0x80208580 0x80000019 mCore | mClassReflector 0x8020ca40 0x80208fa0 0x802085c0 0x8000001a mCore | memoryRegionManager 0x8020c8e0 0x80209000 0x80208600 0x8000001b mCore | sharedMemoryManager 0x8020c780 0x80209060 0x80208640 0x8000001c mCore | mSystemReflector 0x8020c620 0x802090c0 0x80208680 0x8000001d mCore | mAVFaultHandler 0x8020c4c0 0x80209120 0x802086c0 0x8000001e mCore | mAVReflector 0x8020c360 0x80209180 0x80208700 0x8000001f mCore | mAVInit 0x8020c200 0x8020d980 0x80208740 0x80000020 mCore | avManager 0x802107c0 0x8020d920 0x80208780 0x80000021 mCore | eventManager 0x80210660 0x8020d8c0 0x802087c0 0x80000022 mCore | downloader 0x802100e0 0x8020d740 0x802088c0 0x80000029 mSystem | fileSystemManager 0x8020ff80 0x8020d6e0 0x80208900 0x8000002a mDrive | openrBusManager 0x8020fe20 0x8020d680 0x80208940 0x8000002b mDrive | openrBusManager(1) 0x80214a00 0x80214500 0x80208940 0x8000002b mDrive | bmnDriver 0x8020fcc0 0x8020d620 0x80208980 0x8000002c mDrive | bmnDriver(1) 0x80211780 0x8020d2c0 0x80208980 0x8000002c mDrive | fbkDriver 0x8020fb60 0x8020d5c0 0x802089c0 0x8000002d mDrive | cardManager 0x8020fa00 0x8020d560 0x802111c0 0x8000002e mDrive | cardManager(1) 0x80214b60 0x80214560 0x802111c0 0x8000002e mDrive | memoryStickDriver 0x80211fc0 0x8020d500 0x80211180 0x8000002f mDrive | memoryStickDriver(1) 0x802118e0 0x8020d320 0x80211180 0x8000002f mDrive | fatFileSystem 0x80211e60 0x8020d4a0 0x80211140 0x80000030 mCOOP | memoryStickWatcher 0x80211d00 0x8020d440 0x80211100 0x80000031 mDrive | oobjectManager 0x80211ba0 0x8020d3e0 0x802110c0 0x80000032 mCOOP | aperiosClass 0x80211a40 0x8020d380 0x80211080 0x80000033 mClass | opowerManager 0x80210500 0x8020d860 0x80208800 0x80000034 mCOOP | oserviceManager 0x802103a0 0x8020d800 0x80208840 0x80000035 mCOOP | ovirtualRobot 0x80210240 0x8020d7a0 0x80208880 0x80000036 mCOOP | odesignedRobot 0x80211620 0x8020d260 0x80211040 0x80000037 mCOOP | osystemLogger 0x802114c0 0x8020d200 0x80211000 0x80000038 mCOOP | ovirtualRobotComm 0x80211360 0x80214980 0x80210fc0 0x80000039 mCOOP | ovirtualRobotAudioComm 0x80211200 0x80214920 0x80210f80 0x8000003a mCOOP | IPStack 0x80215660 0x80214860 0x80210f40 0x8000003b mDrive | OrinocoDriver 0x80215500 0x80214800 0x80210f00 0x8000003c mDrive | OrinocoDriver(1) 0x802157c0 0x802148c0 0x80210f00 0x8000003c mDrive | OrinocoEnabler 0x802153a0 0x802147a0 0x80210ec0 0x8000003d mDrive | hookConsoleIO 0x80215240 0x80214740 0x80210e80 0x8000003e mDrive | emergencyMonitor 0x802150e0 0x802146e0 0x80210e40 0x8000003f mCOOP | netconf 0x80214f80 0x80214680 0x80210e00 0x80000040 mCOOP | anttcpio 0x80214e20 0x80214620 0x80210dc0 0x80000041 mDrive | hookConsoleIOActivator 0x80214cc0 0x802145c0 0x80210d80 0x80000042 mCOOP | WatchDog 0x802167c0 0x802144a0 0x80210d40 0x80000043 mCOOP | WalkerDemo 0x80216660 0x80214440 0x80210d00 0x80000044 mCOOP | Walker 0x80216500 0x802143e0 0x80210cc0 0x80000045 mCOOP | RoleSelection 0x802163a0 0x80214380 0x80210c80 0x80000046 mCOOP | Recovery 0x80216240 0x80214320 0x80210c40 0x80000047 mCOOP | LaidBack 0x802160e0 0x802142c0 0x80210c00 0x80000048 mCOOP | HlCGen 0x80215f80 0x80214260 0x80210bc0 0x80000049 mCOOP | Hl2Ll 0x80215e20 0x80214200 0x80210b80 0x8000004a mCOOP | CSMonitor 0x80215cc0 0x80217180 0x80210b40 0x8000004b mCOOP | ConfigServer 0x80215b60 0x80217120 0x80210b00 0x8000004c mCOOP | Cogn 0x80215a00 0x802170c0 0x80210ac0 0x8000004d mCOOP | tinyFTPD 0x80217fc0 0x80217060 0x80210a80 0x8000004e mCOOP | powerMonitor 0x80217e60 0x80217000 0x80210a40 0x8000004f mCOOP | | It looks there is something wrong with ovirtualRobot. How can I do with it? | |
| +- 0753 Sorry, I can't edit the post above. ddnt [2003/09/20-02:02] |
| | Response : Edit : Delete |
|
| The
programs failed to run on this robot can run on anyother robots we have. | But the explorer can run on this robot. | |
| +- 0754 RE:Sorry, I can't edit the post above. AiboPet [2003/09/21-01:39] |
| | Response : Edit : Delete |
|
| >The
programs failed to run on this robot can run on anyother robots we have. | ... | > It looks there is something wrong with ovirtualRobot. How can I do with it? | | [general suggestions] | | I've seen similar cases where my program is trashing memory and eventually | when it gets back into OVirtualRobot, then the damage is detected by a crash. | | I assume you are using the 'memprot' system software. Even so, you can still | corrupt system data needed for the robot functions. | | I suggest moving around memory a little. Perhaps by rearranging the OBJECT.CFG | list. | | Also compare the OpenR version (in the Flash ROM) of the offending robot. If | you upgrade to the latest version (using the flash updater) it may fix things, | or just move around memory sufficiently to avoid the problem. | |
| +- 0755 RE:Sorry, I can't edit the post above. ddnt [2003/09/23-00:52] |
| Response : Edit : Delete |
|
Thanks
for your reply. I have tried a program running with nonmemprot mode, there is no exception now, but there was nothing happened at all, no movement as expected, and even no flashing. I can't get connect to the robot. I think there is something wrong with the system on-board, but it is difficult to convince Sony supports in Europe while they insist the robot can performance demo and explorer. I hope the new fireware can solved the problem since I can not update it with current updater. >>The programs failed to run on this robot can run on anyother robots we have. >... >> It looks there is something wrong with ovirtualRobot. How can I do with it? > >[general suggestions] > >I've seen similar cases where my program is trashing memory and eventually >when it gets back into OVirtualRobot, then the damage is detected by a crash. > >I assume you are using the 'memprot' system software. Even so, you can still >corrupt system data needed for the robot functions. > >I suggest moving around memory a little. Perhaps by rearranging the OBJECT.CFG >list. > >Also compare the OpenR version (in the Flash ROM) of the offending robot. If >you upgrade to the latest version (using the flash updater) it may fix things, >or just move around memory sufficiently to avoid the problem. |
| V 0750 how to calculate color pixels smuthu [2003/09/16-14:59] |
| | Response : Email : Edit : Delete |
|
| I am
working with AIBO for a research project and I am using the VCSample 251 | sample codes for it. I actually use AiboCap program that can capture an image | and store it in BMP format in the PC. I need to know if there is a way using | RCODE to calculate the exact number of pixels of any color thats in the image. | Even if AIBO recognizes only color such as pink, blue, red, green ,yellow.. I | need to know if there is a way in RCODE to calcualte the number of pixels of | theses colors in a image??? Also I wanted to know if there is some way you | could define your own global variables (say 5 -6 global variables)in the memory | stick and store some sensory values like distance or color information in them | and then send all those global variable values to the pc at once using RCODE | and VCSample 251 ??? | | Thanks !!! | |
| +- 0751 RE:how to calculate color pixels AiboPet [2003/09/17-02:52] |
| Response : Edit : Delete |
|
>I
am working with AIBO for a research project and I am using the VCSample
251 >sample codes for it. I actually use AiboCap program that can capture an image >and store it in BMP format in the PC. So I assume you are using RCodePlus (2.5x) running on the AIBO. ? I need to know if there is a way using >RCODE to calculate the exact number of pixels of any color thats in the image. Yes, with RCodePlus, use AP_COLORFND (returns pixel count and centroid) http://www.aibohack.com/rcode/rcp_image.htm AP_COLORLVL is faster, but less precise (returns level 0->255 returned by hardware). >Even if AIBO recognizes only color such as pink, blue, red, green ,yellow.. I >need to know if there is a way in RCODE to calcualte the number of pixels of >theses colors in a image??? Correct. Each color range is a different color table entry. AP_COLORFND takes a bitmask, so you can include one or more color ranges easily. > Also I wanted to know if there is some way you >could define your own global variables (say 5 -6 global variables)in the memory >stick and store some sensory values like distance or color information in them >and then send all those global variable values to the pc at once using RCODE >and VCSample 251 ??? RCodePlus supports arrays: http://www.aibohack.com/rcode/rcp_array.htm You can save/load arrays to/from the memory stick. You can also quickly retrieve an entire array's content using the telemetry interface (see TELEMOP_GETARRAYRAW and the 'GetArray' sample) ---- General advice: For RCODE/RcodePlus questions, the AiboLife BBS is a better place to ask: http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi?ubb=forum;f=5 For general RCodePlus extensions, please read: http://www.aibohack.com/rcode/index.html |
| V 0728 mount memory stick read-writer for LINUX j_arroba [2003/09/05-03:27] |
| | Response : Email : Edit : Delete |
|
| Hi, | I want to know how to mount the memory stick read-writer with LINUX. My model is | MSGC-US10. Do i need a driver? How can i know the device where to mount the | filesystem? Thank you very much. | J | |
| |- 0747 RE:mount memory stick read-writer for LINUX lotek [2003/09/12-18:59] |
| | Response : Email : Edit : Delete |
|
| >Hi, | >I want to know how to mount the memory stick read-writer with LINUX. My model is | >MSGC-US10. Do i need a driver? How can i know the device where to mount the | >filesystem? Thank you very much. | >J | | Hi, | I don't know about your model in particular but I think most memorystick readers (if they use | usb) use the usb-storage driver which is included in the standard kernel AFAIK. | See http://www.fredshome.org/Linux/UnoMas/ for a description on how to setup the Unomas | reader, which should very similar to what you have to do. | | Lars | | |
| +- 0748 RE:mount memory stick read-writer for LINUX wangy22 [2003/09/14-10:33] |
| | Response : Edit : Delete |
|
| How
to access Sony AIBO Memorystick using Dazzle USB Memorystick Reader
under | Linux. | | Yu Wang (wangy22@mcmaster.ca) | | INTRODUCTION | | Dazzle Memorystick Reader allows people access information of a memorystick | media, which is mainly produced by Sony, via USB port. This documentation shows | simple steps of accessing the Sony AIBO Memorystick using this read under Linux | operation environment. | | It is tested in Redhat 9 on Acer Travelmate 332t laptop computer. The kernel | version is 2.4.20-8. | | MOUNTING THE MEMORYSTICK | | It is recommended to plug the usb reader into the USB port before turn on the | power of the computer, although most of the linux distributions support hotplug | technology for USB devices. Make sure the Sony AIBO Memorystick is correctly | plugged into the usb reader. | | After log into the bash shell, make sure that the usb reader is recognized by | the system. You can check it by typing: | | $ cat /proc/bus/usb/devices | grep S: | | and you will probably see something like this: | | S: Product=USB OHCI Root Hub | S: SerialNumber=ccacb000 | S: Manufacturer=SCM Microsystems Inc. | S: Product=eUSB MemoryStick Reader | S: SerialNumber=0000000011CF | | Now we can mount the memorystick into the system by the following steps: | | 1. Create a empty directory to which be mounted, ex. /mnt/usb: | $ mkdir /mnt/usb | | 2. And then mount the memorystick (You must have supervisor privilege first) | $ mount /dev/sda1 /mnt/usb | | USING THE MEMORYSTICK | | After mount the memorystick to the specified directory(/mnt/usb), you can use | it just like other storage device, such as harddrive. Most of the filesystem | command, such as ls, mkdir, cat, etc., can be applied. | | When you "ls" the memorystick for the firsttime, you will find a file names | memstick.ind already existing in the memorystick. DO NOT delete it, since this | memorystick is to be used by the Sony AIBO robotic dog. | | If you find any errors or anything can be improved in this HOWTO documentation, | please notify me by email. Thanks! | |
| +- 0749 RE:mount memory stick read-writer for LINUX ejt [2003/09/16-03:32] |
| Response : HomePage : Email : Edit : Delete |
|
>When
you "ls" the memorystick for the firsttime, you will find a file names >memstick.ind already existing in the memorystick. DO NOT delete it, since this >memorystick is to be used by the Sony AIBO robotic dog. 'memstick.ind' is an odd little file, I'm not sure what to make of it. I've seen some places on the web that say this is used by the DRM setup (and hence do-not-delete it) and others that say it's just a flag that is checked by software to determine if a drive is a memstick or some other device (i.e. the software checks for a file named memstick.ind, and if it finds it, assumes the device is a memstick, otherwise doesn't) The latter suggest that if you accidentally delete the file, you can just create a new file with the same name. Personally, i've accidentally deleted the file on several memsticks, and have not found any side effect. I haven't bothered recreating the file since the Aibo doesn't seem to notice. *shrug* Is there any official word on this mysterious little file? The best I've found is: http://www.memorystick.org/mseasy/eng/application/mstickformat.html Which tends to support the simple file-existance-as-flag theory. So I don't pay any attention to taking special care of memstick.ind when i'm moving data around. If someone has evidence that I should, let me know! :) |
| V 0743 I Need A Good Beginning Tutorial..... C3 [2003/09/12-03:57] |
| | Response : HomePage : Email : Edit : Delete |
|
| I am
a complete Newbie to AiBo. My Comp Sci teacher got the school to get us
a | ERS-220 in order for me and a fellow student to fool around with and Program | for our class. I am currently getting everything going and installing (the | OPEN-R SDK) as I am one of two Linux Admins in the Comp Sci computer lab. I | will compile and run the sample programs, but I would like a good tutorial | (that is easy to read) that explains how to write your own programs for AiBo. | My friend and I both know C++, and we both are pretty good at it we think (we | got 5's on the AP Computer Science AB exam we took last May if that | helps....). Thank you very much, and I am looking forward to programming AiBo. | |
| +- 0746 RE:I Need A Good Beginning Tutorial..... grbolar [2003/09/12-10:37] |
| Response : Edit : Delete |
|
Why
don't you try the 'OPEN-R SDK School' tutorials in the 'OPEN-R SDK University (2/2)' page on this web site? |
| V 0741 Inter object communication ejt [2003/09/12-00:24] |
| | Response : HomePage : Email : Edit : Delete |
|
| I'm
passing messages and would like to retain references to them in both the | subject and observer so that I can use it as a shared memory region, in | non-memory protection mode. Part of the trick in this is that each OObject has a | *separate* reference counter. So the subject doesn't know if the observer is | retaining any references, and vice versa. Is this correct? | | Further, the subject is the only one that can release the region? That was the | impression I got from the level2 documentation... only the "owner" can release | it when the refcounter==1, and I'm guessing the owner is the subject. | | Finally is there a way to tell when the memory an RCRegion points to has been | released? I'm trying to track down memory leak/accessing freed memory issues. | | Thanks, | -ethan | (The article that 2003/09/11-13:13 posted, was modified by Author) | |
| +- 0745 RE:Inter object communication ejt [2003/09/12-10:30] |
| Response : HomePage : Email : Edit : Delete |
|
So,
I think I figured out part of the issue. When you have an RCRegion, and pass it to OSubject::SetData(), this will increment the reference counter. When the region is received by the observer, you must then AddReference (if you want to keep it around), which will increment that OObject's counter. Later, when the OObserver's reference counter goes eventually back down to 1, the OSubject's reference counter is then decremented by the system to balance the increment by SetData(). So, in short, when the region is sent, a reference is added for the observer oobject and it starts its own reference counter. When this counter is eventually reset, the observer then removes its reference back from the owner's (subject's) counter. I'm still having a bit of odd behavior with what I'm working on, but I can't put my finger on it. >Finally is there a way to tell when the memory an RCRegion points to has been >released? I'm trying to track down memory leak/accessing freed memory issues. I don't see a way to do this one... not as necessary, but would be nice... -ethan |
| V 0739 Technical diagrams ejt [2003/09/11-23:38] |
| | Response : HomePage : Email : Edit : Delete |
|
| Dear Sony: | It would be great if you could release some of the technical drawings for the | ERS-2xx and the ERS-7 (when it becomes available). It's very hard to do | accurate kinematics and motion models when we have to take measurements | ourselves. The model information PDF's have some lengths, but I would love to | see the moments of inertia, and more detailed lengths. | After all, anyone with a ruler could measure lengths, just not as accurately ;) | But it's especially hard to measure moments of intertia necessary for doing | dynamics! | Thanks! | (The article that 2003/09/11-12:51 posted, was modified by Author) | |
| +- 0740 Author DELETE ***** [2003/09/12-10:05] |
| Response : Edit : Delete |
|
(The
article that 2003/09/11-23:32 posted, was modified by Author) |
| V 0744 code muohioaibo2 [2003/09/12-04:24] |
| Response : Email : Edit : Delete |
|
Can
anyone tell me where I could find code for the aibo dogs. More spacifically, the code for Robocup teams. Any information that would point my team in the right direction would be greatly appreciated. |
| V 0737 Installing OPEN-R SDK in RedHat Linux 8.0 C3 [2003/09/11-11:09] |
| | Response : HomePage : Email : Edit : Delete |
|
| Is
it just me, or is the installation guide for installing hte SDK on
linux a | little inadequate? It pretty much says how to unzip packages...... And that's | about it. I would like to know how I should install it in RedHat Linux 8.0, | and if anyone could tell me or give me a tutorial/how-to that would be great. | Thanks in advance! | |
| +- 0738 RE:Installing OPEN-R SDK in RedHat Linux 8.0 ejt [2003/09/11-12:34] |
| | Response : HomePage : Email : Edit : Delete |
|
| >Is
it just me, or is the installation guide for installing hte SDK on
linux a | >little inadequate? It pretty much says how to unzip packages...... And that's | >about it. | There's isn't much more to it really - you just run the build-devtools script | and it does most of the work. By default, things go in /usr/local/OPEN_R_SDK, | but the OPENRSDK_ROOT environment variable *should* (hopefully) be able to | redirect scripts to a different location if that's inaccessible. | | If you want to see another tutorial, we have one here: | http://www.tekkotsu.org/SDKInstall.html | there's also a very straightforward one here: | http://www.andrew.cmu.edu/course/15-491/homeworks/open_r_setup.pdf | |
| +- 0742 RE:Installing OPEN-R SDK in RedHat Linux 8.0 C3 [2003/09/11-13:36] |
| Response : HomePage : Email : Edit : Delete |
|
>>Is
it just me, or is the installation guide for installing hte SDK on
linux a >>little inadequate? It pretty much says how to unzip packages...... And that's >>about it. >There's isn't much more to it really - you just run the build-devtools script >and it does most of the work. By default, things go in /usr/local/OPEN_R_SDK, >but the OPENRSDK_ROOT environment variable *should* (hopefully) be able to >redirect scripts to a different location if that's inaccessible. > >If you want to see another tutorial, we have one here: >http://www.tekkotsu.org/SDKInstall.html >there's also a very straightforward one here: >http://www.andrew.cmu.edu/course/15-491/homeworks/open_r_setup.pdf ahh that is much more helpful.... it makes sense now. I didn't see the build- devtools because you got to downlaod it seperately. my problem was i didnt see a script or executable to run with the SDK download, and the installation guide didn't say much anyway... Thanks, I think I know what to do now. |
| V 0735 Mac remote montgrimpulo [2003/09/09-17:48] |
| | Response : Email : Edit : Delete |
|
| Hi, | | I have a Mac and an AIBO ERS-210 series. | What Mac hardware do I need if I want to | use the AIBO wireless LAN card with my Mac? | | It should be possible as I have found on | http://www.onrobo.com/reviews/AIBO/ | a Jam.app program which apparently helps | to control an AIBO over a wireless LAN. | | Is an Apple Airport + card sufficient | and are there alternatives? | |
| +- 0736 RE:Mac remote ejt [2003/09/10-04:02] |
| Response : HomePage : Email : Edit : Delete |
|
Ah,
another Mac user ;) You'll like this then... all you need is what you
see here: http://www-2.cs.cmu.edu/~tekkotsu/Samples.html#k-Armed_Bandit Although most of what you need is internal: Airport in the laptop, and a wireless card in the Aibo. In particular, if you want a simple remote control, you can use the tools you see here to get pretty complete telepresence: http://www-2.cs.cmu.edu/~tekkotsu/TekkotsuMon.html#AiboCam_Java They're written in Java, but since my own machine is a Mac and my lab machine is Linux (and some Windows), you can bet the cross-platform compatability is well tested ;) (The article that 2003/09/10-03:52 posted, was modified by Author) |
| V 0733 Regarding memstick reader/writers ejt [2003/09/08-04:11] |
| | Response : Email : Edit : Delete |
|
| I
just thought I should point out that the oft-quoted "Use only Sony brand | readers" warning is more of an extension of "Name brands are more reliable" warning. | | Memory sticks are not a new format at this point - if you buy a reader, you | should be able to expect it's not going to eat your memory sticks. Otherwise, | you should be able to expect the manufacturer to reimburse you for the | damages... assuming the manufacturer is a responsible company, which of course, | not all are. | | We haven't lost a stick yet. We have used a variety of USB reader/writers | without issue, each of which cost around $20. They are: | Model #: (?) from Dazzle (used in Red Hat 7.1, Windows XP, and Mac OS X) | Model #: MSUSB from SmartDisk (used in Red Hat 7.1 and Windows XP, some OS X) | Model #: SDDR-97 (ImageMate MS/MS Pro) from SanDisk (used in Red Hat 7.1) | | The last is a USB 2.0 device... it is slightly faster in linux, but not as much | as I would have expected. It could be USB 2.0 support is lacking in Red Hat | 7.1, we don't have any USB 2 machines with Windows installed to test this. | | Personally, I don't like the idea that *my data* on storage media that *I own* | will suddenly become unusable because of some fault in a system that has no | right to be interfering with my access to my data in the first place. (It may | have a place with commercial software, so fine, put it there; but I don't | appreciate companies who put my investments at risk so they can make a few more $) | Let's think for a second where the blame for faulty and cumbersome DRM (digital | rights management) should be placed. | | Just for argument, even if you *are* using a Sony reader, let's just say you | accidentally pull the stick out while it's being written, which leaves the file | system corrupted. (it's happened) If your disk utils can't fix it, what do you | do? After all, you can't format it because that will clear the "secret bits" | which are used for DRM. Hope you didn't have any precious data on there, | because you now have a little piece of useless plastic. How nice! (luckily, I | *was* able to fix it in our case) | | Complain tho I do, kudos to Sony for allowing free speech on their message board | and making so many great products... I still like them anyway ;) | |
| +- 0734 RE:Regarding memstick reader/writers AiboPet [2003/09/09-01:51] |
| Response : Edit : Delete |
|
BTW:
I generally agree with you sentiments about DRM getting in the way of paying customers. General comments on pink stick rot problems: >I just thought I should point out that the oft-quoted "Use only Sony brand >readers" warning is more of an extension of "Name brands are more reliable" warning. I think most/all pink stick rotting problems can eventually be narrowed down to people fully formatting the stick (sometimes easily fixable), or because of non-Sony brand readers. That's why I recommend Sony Brand, it reduces the chances of a problem [and a stick reader is typically cheaper than the cost of one average pink stick] >Memory sticks are not a new format at this point - if you buy a reader, you >should be able to expect it's not going to eat your memory sticks. Otherwise, >you should be able to expect the manufacturer to reimburse you for the >damages... assuming the manufacturer is a responsible company... SkepticPet asks: Wow, can you name one "responsible company" that would do that ??? In an ideal world agreed, but not in this one ;-> ----- >We haven't lost a stick yet. We have used a variety of USB reader/writers >without issue... BTW: Operating system is also an issue. One $12 stick reader works great under WinXP, but has problems under Win98. I had spotty luck with earlier Linux kernels. ... >Just for argument, even if you *are* using a Sony reader, let's just say you >accidentally pull the stick out while it's being written, which leaves the file >system corrupted. (it's happened) If your disk utils can't fix it, what do you >do? BTW: Usually a format (even unconditional) under WinXP using a good brand reader will fix most ills. BTW2: After adding the stronger warnings on my site - the occurance of this problem has gone *way* down. |
| V 0731 Can't write data into Memory stick youge [2003/09/07-01:38] |
| | Response : Email : Edit : Delete |
|
| hi, | I have an aibo ERS_210A and an Aibo programming Memory stick. | Now the Memory Stick can't be written. I had done the following things. | 1. Copy the system file OPEN_R by Aibo Master Studio Setup. | 2. Save a sample behavior into the Memory Stick. | 3. Put the Memory Stick into Aibo. | 4. Press pause button but Aibo release. Pat head sensor, Aibo has no | response. | 5. I have to press the pause button, Aibo can't pause and the chest light | flash green. I get out the Memory Stick and can't written data in it.(Memory | stick is unlocked) | Why Aibo has no response? | Does the Memory Stick be destroied? | Aibo: ERS-210A | Aibo Master Studio: Aibo Master Studio ERF-PC03 | Memory Stick Adapter: not sony recommended | | |
| +- 0732 RE:Can't write data into Memory stick AiboPet [2003/09/08-02:36] |
| Response : Edit : Delete |
|
>hi, > I have an aibo ERS_210A and an Aibo programming Memory stick. >Now the Memory Stick can't be written. I had done the following things.... > Why Aibo has no response? > Does the Memory Stick be destroied? ... I don't fully understand your problem. First of all, using a non-Sony brand stick reader is dangerous for anything - it can wipe out the copy protection making the stick unusable for AIBO. If AIBO boots up with the memory stick in it, even if AIBO doesn't do anything - you are ok. [if however it shuts down after 5 seconds with a 5 note "duh duh duh duh DUHHH" sound, you may have broken the stick] ---- Suggestions: Reboot your computer, insert the memory stick to the PC reader, and see if you can delete all the files from the stick [DO NOT FORMAT] If that doesn't work, I would suggest buying a Sony brand stick reader. If that does work, I would install a sample program to see if it is working. For example, DogsLife http://www.aibohack.com/dogslife If that works, delete the stick contents again, and go back to the AMS setup and try again. ======= If you have continued problems, please post RCODE/AMS questions on the public BBSs [you will get better and faster responses] http://www.aibo-life.org/forums/cgi-bin/ultimatebb.cgi?ubb=forum;f=5 Good luck |
| V 0729 Mem Stick problems?? aung [2003/09/06-13:28] |
| | Response : Edit : Delete |
|
| Hi,
Aibo does not boot up after installing memory stick. It makes a sound
and | the chest light goes off. I tested with Aibo-life and then it works. So, Aibo | side seems to be able to load mem stick data. | | I tested with 2 of my memory sticks. They are still readable and writeable in | PC side. What can go wrong? | | Checked the Bat. 3 levels left. | | It was working fine before | | Rgds, | Aung | |
| +- 0730 RE:Mem Stick problems?? AiboPet [2003/09/07-00:33] |
| Response : Edit : Delete |
|
>Hi,
Aibo does not boot up after installing memory stick. It makes a sound
and >the chest light goes off. ... [and related general pink stick advice] I assume the bad memory sticks are PMS/RCODE sticks for use with the OpenR SDK. First make sure you have a correct program installed on the memory stick. Any previously working OpenR SDK or RCODE program will work as a test. I use my Poser program (small and fast OpenR SDK program) for such tests http://www.aibohack.com/openr_sdk/index.html#POSER --- If you still have problems, and hear a 5 note "duh duh duh duh DUHH" sound approximately 3 to 5 seconds after pressing the button, then possibly/likely the copy protection of the pink stick is corrupted. The number one reason for this is using a non-Sony brand memory stick reader. WARNING: ONLY USE SONY BRAND MEMORY STICK READERS! WARNING: NEVER FORMAT A PINK MEMORY STICK! |
| V 0727 OPEN-R Techno Forum in Fukuoka, Japan openrsupport@(Administrator) [2003/09/04-14:30] |
| Response : Edit : Delete |
|
The second "OPEN-R Techno Forum" was recently held in Fukuoka, Japan. One of the AIBO Owners who attended that event has graciously allowed us to link to his website, which contains pictures and movies of the event. The website is http://www.photohighway.co.jp/tp/14_f.asp?key=1371486&un=25215&m=0&pa=&Type=14 Thank you, OPEN-R SDK Support |