Micromouse - the early years, 1980
Practical Mechanics August 1980
New Scientist 25 September 1980
Micromouse is a good mover but lacks breakthrough
Eight microprocessors on wheels raced to thread a maze last week. And all but one got hopelessly lost or stuck. The European Micromouse Maze Contest, part of the Euromicro conference at Imperial College, London, was supposed to present a challenge for Europe's brightest young engineers. It did indeed reveal their strengths and weaknesses. The only consolation remaining is that the Japanese entry did not even reach the starting line.
The $1000 winner. Sterling Mouse, built in West London by an electrician and a British Gas project analyst, out-performed entries from universities and well-known electronics firms. Fred, for instance, built in spare time by two Plessey engineers, had some very professional-looking printed circuit boards under his furry cover. But his builders admitted he was running "very unintelligent software". The ICL Amateur Computing Club's mouse had some very clever software. But it was held together by yellow PVC tape, and had to retire with trouble caused by a dry solder joint and a broken wire.
Meryl, bult by staff of Marconi Research Laboratories, seemed almost paralysed and needed constant shoves (though she looked decidedly livelier on BBC TV's Blue Peter the following day).
A Swiss entrant, whose mouse was built with a watchmaker's precision, complained that the maze itself was not built precisely enough. As if to underline the point, a toy-car style entry from Lancaster University got bogged down attempting multi-point turns in narrow
Some mice bristled with sonars and photocells. But the home-built winner used two bits of bent-brass shim and half a dozen bare metal contacts to feel the maze walls. It worked almost perfectly, and like all good engineering it looked simpler than it was.
Two mice which failed to negotiate the maze nevertheless shared the prize in the Virtuoso Display section. Fred danced the Blue Danube Waltz to his own accompaniment; Midnight Sun from Finland wrote its name and then played a Beatles' tune while gyrating on the spot. Neither of these preprogrammed displays, of course, required an ounce of
Actually, entrants admitted that even a maze-solving program is not hard to write. "Anybody can write a maze-solving algorithm in a day," said Phil Yeardley, whose mouse was built from Lego bricks. The Lego mouse ran well until one of its feelers got stuck in a piece of sticky tape. "The biggest problem is steering," said one of the Finnish team, confirming that the mechanical problems are the toughest.
Perhaps there is a valuable lesson here. It is no use training software wizards unless we also train electrical and mechanical engineers who can give the chips the eyes and feelers, the hands and the wheels they need to do their job.
A fight for the Finnish: a very unlikely-looking "mouse" from Field Oy, Helsinki, did not reach the centre of the maze, despite threats . . .and blandishments.
21st October, 1980.
Thank you for your interest in the Micromouse Maze Contest. I enclose a copy of the rules, together with an entry form which I hope you will return as soon as possible. In the late spring-local heats may be organised, and I hope that prizes will take the form of sponsorship for the Paris finals.
If you have any problems, please do not hesitate to get in touch - my home telephone number is (0705) 812466.
Dr. John Billingsley
P.S. A British heat will be held at the On-Line Exhibition in the Wembley Conference Centre at the end of July.
Microprocessor controlled robot mice must find their way to the centre of the
The maze consists of 16 x 16 squares. The squares are based on a
7 inch/18 cm matrix. The walls of the maze are 1/2 inch/12 mm thick, and
the passageways are thus 6 1/2 inch/16.5 cm wide. The walls are 2 inch/5 cm
high, painted white with red tops. The target post at the centre,
1 inch/2.5 cm square, is 8 inches/20 cm high, and can be removed if desired.
The starting square is at the 'bottom left' corner of the maze, and the
mouse is initially oriented so that the target is diagonally to its right.
The running surface is chipboard, sprayed with black paint.
Dimensions should not be assumed to be more accurate than 5%: the
maze may be made to metric or imperial dimensions, and quoted figures may
be approximations (to 5%). Joins in the maze base will not involve steps
of greater than O.5 mm - possibly covered with tape. However, warping of
the maze base during transport or storage may result in a change in
gradient at a join of as much as 4 degrees.
3. Contest rules
(a) Each mouse is allowed a maximum total of 15 minutes to perform. The
judges have the discretion to request a mouse to retire early if by its
lack of progress it has become boring, or if by erratic behaviour it is
endangering the state of the maze.
(b) If the mouse can succeed in finding its way from the start to the
maze centre, the time is noted. The handlers can then restart the mouse,
so that it can profit from any learning ability in making a second run.
Within the 15 minute limitation any number of runs can be made, and the
mouse is credited with the shortest time of a successful run.
- 2 -
(c) If a mouse 'gets into trouble', the handlers can ask the judges for
permission to abandon the run and restart the mouse at the beginning. If
any other handling occurs, the judges may impose an appropriate time penalty,
and a mouse which has been so handled will be beaten by any other mouse
which has reached the centre without handling. A mouse may not be re-
started merely because it has taken a wrong turning - the judges decision
is final. The judges may add a time penalty for a restart.
(d) Within reason, and at the judges' discretion, battery changes and
minor repairs may be made - however the 15 minute timer must continue
(e) If no successful run has been made, the judges will make a
qualitative assessment of the mouse's performance, based on distance
achieved, 'purposefulness' versus random behaviour and quality of control.
(f) If a mouse elects to retire before three minutes have elapsed, the
judges may at their discretion permit it to perform again, with a 1O
minute limit, later in the contest.
(g) The judges will use their discretion to award the prizes, which in
addition to the major prize may include prizes for specific classes of
mouse - perhaps lowest cost, most ingenious, best presented, etc.
(h) Before the maze is unveiled the mice must be accepted and caged by
the contest officials. The handlers will place the mice at the start
under the officials' instructions.
(i) Although the superstructure of the mice may 'bulge' above the top
of the maze walls, mice must be subject to the following size constraints -
width 25 cm, length 25 cm. There is no height limit but beware of
toppling! Mice must be completely self-contained and must receive no
outside assistance. The method of wall sensing is at the discretion of
of the builder, however, the mouse must not exert a force on any wall of
greater than one Newton. The method of propulsion is at the discretion of
the builder provided that the power source is non-polluting - internal
combustion engines would probably be disqualified on this count. If the
judges consider that a mouse has a high risk of damaging or sullying the
maze they will not permit it to run. Nothing may be deposited in the maze.
- 3 -
The mouse must negotiate the maze; it must not step over or otherwise
illegally cross any maze wall. The means of locomotion of the mouse is
again at the discretion of the designer.
4. Virtuoso Display
Mice have 10 minutes to display their abilities. Their dimensions
must be such that they could run in the maze. Their performance must be
limited in travelling range to be accommodated either within the maze, or
on an 8 foot square (2.4 metre) surface of hardboard. Handling will be
The Micromouse Maze Contest was first held in the U.S.A. by IEEE Spectrum.
||MICROMOUSE MAZE CONTEST|
PARIS 7-10 SEPTEMBER 1981