Subject:
|
Re: MIndstorms Neural Network in Java
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Fri, 2 Nov 2001 22:16:06 GMT
|
Original-From:
|
David McCafferty <dave@saab.demon.&NoSpam&co.uk>
|
Reply-To:
|
DAVE@SAAB.DEMON.CO.UKsaynotospam
|
Viewed:
|
773 times
|
| |
|
|
I'm going to get the robot to navigate around a maze, which is gettin
built on sunday so then i'll have a better idea of what i am dealing
with exactly as its not going to be the classic type maze that i could
just follow a wall as far as i know. But the idea is that it would
"learn" a bit using the neural net and this would hopefully aid it to
complete the task faster.
I will also have to tie in the other sensors and webcam, the webcam
will be to pick up visual hints and also look for obsticles!
it could prove to be quite complicated i am guessing :(
I do realise that it would probably be quicker in c/c++ however i am
being strongly advised to do it in Java and as a bonus it will get me
out of doing an assignment for the Java module I am taking which could
prove handy! Also the advantage of Java is that it was origionally
written to program embedded systems so in a way a mindstorms project
is perfect for it.
thanks for your comments any other help would be greatly appreciated
Thanks
David
_
Don't Look back unless you can Laugh.
Dont Look Forward Unless you can Dream.
|
|
Message is in Reply To:
 | | Re: MIndstorms Neural Network in Java
|
| (...) Neural nets use a LOT of CPU time - you might want to consider using C code with LegOS - it's quite a bit faster than Java. What kind of application did you have in mind? ---...--- Steve Baker ---...--- Mail : <sjbaker1@airmail.net> WorkMail: (...) (23 years ago, 2-Nov-01, to lugnet.robotics)
|
3 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
