Subject:
|
SMART's 2004 Grand Challenge: Robot Soccer
|
Newsgroups:
|
lugnet.org.us.smart
|
Date:
|
Thu, 5 Feb 2004 01:41:54 GMT
|
Viewed:
|
1903 times
|
| |
| |
At our last meeting we voted on a new Grand Challenge: Robot Soccer. This will
be an ongoing challenge for 2004. The reason for a Grand Challenge is that this
lets you build a robot over a longer period of time. You will have many
opportunities to try it out, make improvements, learn what others are doing so
that you can build something even better, etc.
I'll bring an IR ball and a mat to every meeting this year. The idea is to build
a robot to score goals against an opponent. The rules we'll use are NOT the same
as the Robocup Jr. soccer rules.
The field will be a 4'x8' paper mat (so don't build robots that will not work on
paper - stay away from treads!) that has a gradient across it -- white at one
end, and black at the other end. There will be a short wall around the perimeter
(probably we'll just use 2x2's) with openings for the goal. We'll start with a
goal that is 24" wide. Over time we'll reduce this to 12" wide.
The IR ball is around 3" in diameter. I think a tennis ball is around 2 1/2" in
diameter, so that's a close substitute at home.
Your robot will start in it's own end, (so look down to see where you're
starting!) and the ball will start roughly in the center of the field. You'll
need to find the ball, and try to roll it into your opponent's goal, avoiding
your opponent.
A match will last five minutes. Perhaps early one we'll do 3 minute matches,
though.
I recommend building robots that won't 'catch' on your opponent -- don't have a
lot of loose beams sticking out. I also recommend that robots figure out if they
are stuck, and if so, they try backing away.
Try to limit the size of your robot as much as possible. We'll come up with some
suitable size limits after a few meetings (probably 10"x10", or so, but perhaps
smaller). Also, the ball must always be "outside" of your robot. You can put a
concave scoop in front, but at a minimum, half the ball should always be outside
of the convex hull of your robot. Your robot shouldn't hold onto the ball.
If you'd like to purchase a ball or mat yourself, the ones I'm using are from
Acroname. The ball is fairly expensive, but the mat is quite reasonable.
<http://www.acroname.com/robotics/parts/R194-ROBO-BALL.html>
<http://www.acroname.com/robotics/parts/R204-ROBOCUPJR-MAT.html>
(Here's another place that makes IR balls, though they're not in the US. This is
not identical to the one I have, but it should be close:
<http://www.wiltronics.com.au/catalogue/shop.php?cid=339>)
One way you can test your robot's ball-finding ability is by using a lightbulb,
(just don't let your robot hit it!). There are many other means. Be imaginative!
As you can see, the rules are pretty loose at this point. The main point isn't
to build the winning entry, though the competitive aspect *is* important. The
idea is to learn how to build a robot that does something complicated, and also
how to improve your robot over time, both mechanically and in its programmed
strategies against other robots.
--
David Schilling
|
|
Message has 2 Replies: | | Re: SMART's 2004 Grand Challenge: Robot Soccer
|
| (...) *SNIP* David, I am glad to see that this challenge is being tackled by another LEGO based robotics group. I have to say that so far this has been the most challenging event we have tackled. (...) *SNIP* We found that robots that get stuck on (...) (21 years ago, 11-Feb-04, to lugnet.org.us.smart)
|
3 Messages in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|