Subject:
|
Re: 3T
|
Newsgroups:
|
lugnet.org.us.laflrc
|
Date:
|
Tue, 3 Jan 2006 17:33:52 GMT
|
Viewed:
|
1326 times
|
| |
|
|
In lugnet.org.us.laflrc, Steve Hassenplug wrote:
> I'm not exactly sure the size, but it was smaller than the others.
> I think it was the only one that was attached to only a 32x32 baseplate.
Yes, Steve's was the smallest, and fastest, and the only one that was
completely on a 32x32 (and a road baseplate at that). It brought to mind a black
cubistic turtle, and was the only one of the three that attatched to the board
on the side (instead of opposite the opponent). Kirby had a beautiful (red),
delicate (in mechanism, not strength) complex that dropped blocks from a
curiously slender articulated "hand", reloading the hand from an angled feeder
tray. John's had a wonderful looking robot, the only one that could reach over
the board without resting upon it. To my eye it recalled an industrial machine
designed for foundry work. Mine was a gantry crane that used a single motor to
drive a pick-and-place EOAT that could retieve the cubes once the game was done
(actually, it could retieve it's cubes or the opponents cubes, to allow for
switch sides).
> It carries five cubes, checks three cells, and was somewhat fast.
Steve is too modest. His is also the only 3T robot of the day that could wipe
out a hoard of attacking Viking minifigs (yes, I have pictures).
> I'm not sure the exact time per move, because it didn't make that
> many moves.
The average game-time for my robot was around a minute, so that's about 20
sec per move. With some more coding, I can push that lower, but I think I can
only get to an average time-per-move of 10 seconds with my current design. *BUT*
Kirby and I were the first people to watch two LEGO robots play 3T without a
single touch from a human... which was very cool. After we played something like
five games, it occured to us that each game was the same, at which point I spent
the rest of my time sacrificing a slight amount of speed for some variety.
> One interesting note is that everyone was using different software
> to program their robots.
I was going to comment on that as well. Yes, I was using NQC. This is also
only the second project that bumped me up against the variable limit of the
standard firmware, and board evalutation took too long for me (3-4 seconds). I
was the only person who did not have an "opening book" in the code as well
(again, I could do this to speed it up... but my mechanism is just nowhere fast
enough to take top honors as it stands).
> I think I'm going to adopt one of Brian's suggestions and make
> it possible for my robot to lose a game, in a rare random situation.
I put that in there (as well as a "human friendly" switch for playing against
kids), and I might make it cheat or very very rare occassions, in a way that
kids can see... ;-).
The best thing about the day is how dead wrong (again) I was. I thought there
would be some common elements betwen all the robots. The only thing they had in
common was placing cubes. With only four robots playing, we had 'bots that would
drop, roll or place their cubes... would scan one or multiple cells at a time...
used gear racks, or string drives, or pneumatics... one or more than one RCX (I
suggested a purely mechanical one, but not yet ;-)... free-hanging arms,
gantries, entirely mobile robots... four different programming enviroments...
four different ways of generating the next move... different ways of attatching
the the board... and four completely different ways of detecting cubes (pushing
cams, light sensors, long touch-sensor "fingers", "whisker" systems read by a
light sensor). Just amazing diveresity.
Um, I'd be lying if I said there wasn't also a lot of frustration. At one
time or another, nearly all the robots tried to "stack" cubes for one reason or
another, and mine lost a game when after 2 moves it proudly announced that it
had just won the game. I never did figure that one out.
--
Brian Davis
|
|
Message has 1 Reply:
 | | Re: 3T
|
| (...) My robot also basically rests on a 32x32 plate...I have an extension off the back more for some stability than anything else. (...) Good news - I also have a low riding front end and can help Steve outflank the Vikings at the next get (...) (18 years ago, 7-Jan-06, to lugnet.org.us.laflrc)
|
Message is in Reply To:
| | Re: 3T
|
| (...) I'm not exactly sure the size, but it was smaller than the others. I think it was the only one that was attached to only a 32x32 baseplate. (...) It carries five cubes, checks three cells, and was somewhat fast. I'm not sure the exact time per (...) (18 years ago, 3-Jan-06, to lugnet.org.us.laflrc)
|
12 Messages in This Thread:
 
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
