mwtilden@watmath.waterloo.edu (M.W.Tilden, Hardware) (10/23/90)
The North American Micromouse championships were held on
Oct 13 and 14 at the Ontario Science Center in Toronto, Canada.
The competiton this year was sponsored by the IEEE and the University
of Waterloo which provided the maze platform.
For those that don't know, the micromouse competition, now into
it's eleventh year, is an open competition to build a robot mouse which
can get from the outside of a 16x16 maze to the middle in the
shortest possible time. North american competitors range from
high school and University students (usually using the MAPPY kit mouse
from NAMECO) to robot hobbyists with some *really* neat toys.
Competitions range themselves from the local to the national to the
international (the Japanese, for example, are nuts on the game).
The North American championships for this year featured seven mice
from 5 teams. The overall winner was MITEE 6 from MIT with a fastest
run time of 13.8 seconds.
The teams, in order of finishing were:
MICE: MITEE 6 and MITEE 3
TEAM NAME: MIT Mouse.
ORGANIZATION: MIT Electrical Engineering.
MEMBERS: Dave Ottman, Jimmy Ottman (age 6-3/4) and Tony Telegero
MICE: MOUSEMOBILE 1 and MOUSEMOBILE 2
TEAM NAME: -
ORGANIZATION: University of Montreal (Canada)
MEMBERS: Martin P'Otte and Louis Je'Fwa.
MICE: RATFINK 1
TEAM NAME: The Ratfinks
ORGANIZATION: California State University (Long Beach)
MEMBERS: Kent Sweeny with Ray, Takashi, Paul, Dora, Chris and
Ron on video.
MICE: HDL 1.00 and Solaroller 2.0 3 2
TEAM NAME: The Mild-Mannered Mad Scientists Society (M S )
ORGANIZATION: Hardware Design, University of Waterloo
MEMBERS: Mark Tilden
MICE: ZETA 3
TEAM NAME: Team Zeta
ORGANIZATION: Orange County, La Palma CA
MEMBERS: Harjid Sing and Jim Rawthorn
The competition went very well with MIT taking home honors and first,
University of Montreal taking home the Canadian Championship, and the
Ratfinks taking home the MAPPY kit mouse trophy (they were the only
MAPPY entry). Most original mouse went to Waterloo for making a
mouse which didn't rotate, didn't use optical sensors and didn't
use a processor (the first ever to do so). Waterloo also took
home recognition for the cheapest mouse ($41.00 including batteries.
Compared to $12,000.00 for MITEE 6) and the slowest mouse (a solar powered
unit which would solve the maze in as little as a half year).
The actual runs were very impressive. MITEE 3 (the current world
champion) did not do as well as it's younger brother MITEE 6 because the
maze floor was a little slicker than the team was used to. MITEE 3's
over the wall sensors added significantly to it's upper body inertia
and thus had to run slower to maintain floor traction. MITEE 6 was very
impressive with it's chipmunk-like darting from position to position
during it's mapping phase. It's final run was facinating as the mouse,
the maze now memorized, cut corners to get to the center in record time.
MITEE 6 featured side mounted optical sensors to find walls but because
it had no over-the-wall sensors took marginally longer to map the maze.
As the maze this year was rather inordinately difficult and long, the
low, sleek MITEE 6 was a shoe-in.
The Montreal mice were the only other competitors to actually make the
middle of the maze. Following adjustment of the lighting conditions
so the optical sensors could make out the maze walls properly, Mousemobile
2 beat out it's older stepper-motor version by over 10 seconds.
The other competitors did not finish due to various software/hardware
problems. For example, the Ratfink mouse kept getting caught up on a
.5mm crack on the maze floor, and the Waterloo mouse could not seem
to figure it's way out of tee-junctions. There's no such thing
as ideal conditions.
Micromousing could be considered to be the Nascar Circuit of the
robot world. As such it was disappointing to see just how few
competitors there actually were. As the Waterloo entry proved, you
don't need lots of money or development tools, just an idea for a
high speed mouse that falls within the rules. Typical prize money
is around $1,000 a shot. With over 4 national and international
competitions a year, the MIT team is pretty much breaking even
with their continual wins. Money is a small incentive compared
to discussing ideas with people who always seem to solve the problem
in different ways. Competitors are extreemly open about their design,
program, algorithms and component sources. It's a great 'meeting of minds'
forum and I urge anyone who can enter to enter. The next North American
competition is in Dallas at the end of Febuary and the next Canadian
competition at the end of March. Before long, videos of this last
competition plus "how to" documentaries will be available so you too
can get your start in micromousing. Pending interest, I'll post when
they'll be available.
For more information on Micromousing (rules, past mazes and times,
etc.), contact the North American Micromouse Association (NAMA) at:
MICROMOM
aka. Sue Rosenbaum
1086 Central Ave.
Plainfield, NJ
07061
USA
Or call (201)757-6749. $25 will get you details plus the semi-occasional
newsletter.
I'd like to point out that now that Waterloo University has a maze to
work on, we're hoping to see many entries over the coming years. And now
that we have videos to study (including close-ups of the competitors
hardware) we can guage ourselves against international competition.
Hopefully.
Is all.
--
Mark Tilden: _-_-_-__--__--_ /(glitch!) M.F.C.F Hardware Design Lab.
-_-___ | \ /\/ U of Waterloo. Ont. Can, N2L-3G1
|__-_-_-| \/ (519) - 885 - 1211 ext.2454,
"MY OPINIONS, YOU HEAR!? MINE! MINE! MINE! MINE! MINE! AH HAHAHAHAHAHAHAHAHA!!"