 | | Re: Very strange problems —Luis Villa
|
| | | (...) As far as I can tell, legOS has some really strange bugs. I've learned to take it for granted that some programs (without warning) will just crash and lock up the machine. This is unfortunate, but I've never found a pattern to them so there is (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | |
| | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) That is really bad news. Since we are writing a genetic programming system with six RCXs involved that communicate with each other, the programs tend to get pretty long. Often things fail. Right now I have the problem that if I save a program (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Luis Villa
|
| | | | (...) Yes. We ran into similar problems when writing neural network code. Persist- you can write really cool stuff on legOS- we did. Also, emulegos (which is still around somewhere, though I'm not sure if anyone has ported it to 0.2.x) would (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) Is this neural network code made public? I have emulegos, but I havn't much more than started it up. Do you know if it can handle IR-communication between several RCXs? Tobias (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Luis Villa
|
| | | | (...) I will hopefully clean up and release some of it by Wednesday. (...) I'm not sure- I don't think so, though. Someone else here may be able to answer that much better than I can. Luis ---...--- "Going to California with an aching in my heart. (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) It seems that emulegos should be able to handle IR. However, it does not use the lnp-functions that I use, so I presume it is designed to work with an earlier version of legOS. Were earlier versions of legOS (I use version 0.2.3) less likely (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Luis Villa
|
| | | | (...) IIRC, it is 0.1.7, which used a completely different setup for the IR communication. (...) Unfortunately, no. (...) You mean, unmodified? No, I don't think. I've seen lots of code that worked, and then got "simplified" and then stopped (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) I was afraid so. (...) We haven't reached the GA part yet. Right now the idea is to work out a communication protocol and then send a information back and fourth. And since the RCXs locked up we thought that there must be something wrong in (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Luis Villa
|
| | | | (...) Argh. I was getting emulegos (which I was pretty familiar with) and legosim (which I though was C, not Java) confused. I'm afraid neither of them would be too easy to port to 0.2.x :( (...) Please make sure to share this with us! IR continues (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) We have not done anything to ensure that whatever is transmitted is received correctly (i.e. without errors). What we have done is a build a system which will work even if one (or several) of the RCXs are temporarily blocked for some reason. (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Luis Villa
|
| | | | (...) Redundancy is good :) (...) Good. It's generally solid, but can cause some issues from time to time. <snip lots of monkey talk :) > (...) Distance of movement or consistency (i.e. always forward) or both for fitness? Sounds really cool. Good (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Michael Ash
|
| | | | | (...) What sort of "issues" are these? Are you referring to the normal issues of multithreading (synchronization, deadlock, etc.), or specific legOS-related problems with it? I have a program that uses a decent amount of multithreading, and I'd hate (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | | |
| | | | | | | Re: Very strange problems —Luis Villa
|
| | | | | | (...) Just that it is not pre-emptive. If you don't msleep() at the appropriate places in your code, you'll never go back to the scheduler or the other threads. This can come as a surprise, since you'll only notice it when one thread never executes. (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | | | |
| | | | | | | | Re: Very strange problems —Michael Ash
|
| | | | | | (...) Ack! I knew that it was cooperative in the 0.1.x series, but I had thought it was preemptive in 0.2.x. Drat. Well, I don't think it's a major problem for me, since I call msleep() religiously, but it's something to look into. Thanks. -- "Our (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | | | |
| | | | | | | | Re: Very strange problems —Luis Villa
|
| | | | | | (...) I have not done large-scale "production" work with 0.2.x, really, so I may be vastly incorrect. Ah-hah. The legOS homepage (at least) indicates that you are correct. My fault for scaring people... sorry. Luis ---...--- "Going to California (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | | | |
| | | | | | | | Re: Very strange problems —Michael Ash
|
| | | | | | (...) Ah, that's all right, happens to the best of us. :) I'll go back to my happy little preemptive world. -- "Our doubts are traitors, and make us lose the good we oft might win by fearing to attempt." - William Shakespeare Mike Ash - (URL), (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | | | |
| | | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) Probably movement since it is easier. Our fitness function may have to change once we see the result though. GA tends to give you individuals that cheat every chance they get :). (...) It will be found at (URL) Right now we have our (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Luis Villa
|
| | | | (...) Neural nets do similar things. Our class had some very interesting cheaters. I actually thought the students learned as much from watching their robots cheat as they did from any other single component, since figuring out the "cheating" forced (...) (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
| | | | | | Re: Very strange problems —Tobias Bende
|
| | | | (...) That seems very likely. (...) Great :). The complete rules for the tournament can be found on the course web page, which is linked from our page. Tobias (24 years ago, 21-May-00, to lugnet.robotics.rcx.legos)
|
| | | | |
