mwtilden@watmath.waterloo.edu (M.W.Tilden, Hardware) (11/03/90)
The following is a paraphrase of two articles sent to me from
Paul Grant of the Turing Institiute (paul@turing.ac.uk). They
discuss details and results of the first International Robot Olympics
held in Scotland. The competition was was organised by Paul and
Peter Mowforth of the Turing Institute (plus others too numerous to
mention). He has tried posting these but without luck, so I'm doing
it now.
-----
The 1st International Robot Olympics took place at the Turing
Institute on the 27 and 28th of September, 1990. The following is a
list of this years winners.
OBSTACLE AVOIDANCE: Asterix (Univ. of Toronto, Canada)
WALL FOLLOWING: Yamabico (Univ. of Tsukuba, Japan)
WALL CLIMBING: Zig-Zag (Portsmouth Polytechnic, England)
POLE BALANCING: Salford Pendulum (Univ. of Salford, England)
BIPED RACE: Cardiff Biped (Univ. of Cardiff, Wales)
PHOTOTROPIC: Alpha Photon (Univ. of Kent, England)
MANIPULATORS: Belgrade/USC Hand (Univ. of Belgrade, Yugoslavia)
TALKING: Richard I (Turing Institute, Scotland)
JAVELIN: York Arher (Museum of Automata, England)
MULTI-LEGGED RACE: Penelope (Univ. of Edinburgh, Scotland)
BEHAVIOUR: Genghis (MIT, USA)
Events were marked by six judges. Generally, robots were awarded
marks based on sophistication of behaviour, novely of design
and quality of the hardware.
A number of special prizes were awarded by the sponsors:
IEEE Robotics and Automation Society Young Roboticist Award:
Brian Carr, St. Patrick's High School, Scotland
Robot: BC1
NatWest Bank Transfer for Technology Transfer:
Massachusetts Institute of Technology (AI Lab), USA
Robots: Attila and Genghis
Best School Award:
Inverkeithing Primary School, Scotland
Robot: Xybot
Overall Olympic Champion:
University of Tsukuba, Japan
Robot: Yamabico
ROBOLYMPICS: RULES AND GUIDELINES.
Firstly, there are no official rules for the each of the competitions.
In fact, there were no rules as such. Each competitor was sent an
application form which asked the type of behaviour their robot could
exhibit. Robots with similar capabilities were then pitted against
each other. The marking criterea was decided by the judges. Some of
the events (wall climbing, walking) were simple first past the post
races. Others were more qualitative in nature.
In the obstacle avoidance the robots were given a square pen (5mx5m max)
and asked to wander around it for two minutes. Objects (non-planar
included) were placed randomly in the pen. Robots were penalised
for collisions and slow speed.
In wall following, robots were given a corridor and asked to navigate
along it. The corridor included turns. Robots were marked on
accuracy and speed.
The pole balancing was judged on a number of qualitative mertics such as
the adaptibility to environment changes (external forces etc).
The phototropic robots were marked on the quality of their behaviour
in following/fleeing light sources.
The manipulators were asked to pick and place objects of various
geometries. These included balls, cups etc. The manipulators where
judges on their accuracy/speed.
The talking robots were asked to repeat utterances. The judges then marked
the robots on pronunciation. Words with silent letters (eg psalm, knife etc)
were included.
The javelin was marked solely on the distance the javelin was thrown.
The behaviour category was the most general event. Robots were simply
asked to demonstrate their behaviour. The judges marked on the quality
of behaviour.
It is important to point we purposely didn't show the exact requirements
for each event to the competitors before the competition. We didn't
want solutions to be engineered, rather we wanted to test how robots
would perform in a general environment. This is in contrast to something
like the micromouse competition where the mice perform in a very
constrained environment.
One of the main aims of the Olympics was to get more people interested
in the practical side of robotics (rather than simply reading Sci Fi books).
Thus, the first day of the Olympics was dedicated to schoolkids. Over
1,000 schoolkids visited the Olympics and all attended talks on
robots before wandering around the arena and seeing the competition
robots. We also had a number of school entries (all from the UK
unfortunately) which were of suprisingly high standard.
Another aim was to get the media interested in robotics. This was probably
the major success of the Olympics and I hope we have increased the profile
of robotics worldwide. We also wanted to highlight the lack of funding
for robotics (and science in general) in the UK.
We also held a robotics masterclass in conjunction with the Olympics.
The speakers were:
Peter Mowforth (Turing Institute, UK)
Hans Moravec (Carnegie Mellon University, USA)
Martin Snaith (UK legged locomotion group, UK)
Ruzena Bajcsy (University of Pennsylvania, USA)
Valerie Gradetsky (USSR academy of sciences, USSR)
An official Robot Olympics Brochure is available and can be forwarded
if you send Paul your full postal address, otherwise you can write or
E-mail Paul for details and discussion at:
Paul Grant,
The Turing Institute,
George House,
36 North Hanover St,
Glasgow G1 2AD,
UK
e-mail: paul@turing.ac.uk
fax: +41 552 2985
------
Rumor has it that the next Robolympics will be held in Japan in
two years time. See you there.
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!!"