MINOS12 Micromouse Contest

By | April 22, 2012

As usual, on the Sunday after the Minos presentations, we hold a semi-formal micromouse contest as a bit of a taster for the upcoming UK summer contest. All the usual suspects were present but for some, this was a difficult year.

I have probably not been to a micromouse contest where there has not been some fuss over the walls. This year, however, the Royal holloway maze seemed particularly challenging. Although the common Korean walls were used, they seem to have come from several different batches and everyone notices the apparently large variations in reflectivity. While this was apparent at visible wavelengths, I can only suppose it was worse at IR wavelengths. there was much shaking of heads and crashing of mice. Here is a picture of part of the maze illustrating some of the problem:

Minos 2012 micromouse maze walls very variable

By the time we got to the finals stage, pretty well nobody had any idea what was going to happen. the maze had two main alternative routes with a number of minor variations possible. Generally mice took the slightly shorter but more twisty route shown on the left. Here is the maze that was used:

Minos 2012 micromouse finals maze route 1 

All in all, it is fair to say that many mice failed to give of their best. Not all though. Since we had no automatic timing mechanism this year and were also lacking anything to automatically calculate search time penalties, we adopted Japan Rules. That is, each mouse was allowed five runs or five minutes – whichever finished first. Touch penalties were not used and a simple fastest run time was the recorded criterion. Here are the results:

MouseOwnerRun 1Run 2Run 3Run 4Run 5Pos
FatCatTim Foden40.811.613.2RR1
PicOne TurboJim Chidley41RRRR2
Decimus 2BPeter Harrison45RRRR3
FAB 1Derek Hall63RR63R4
EraticDavid Hannaford6870---5
PicOneJim ChidleyRR104RR6
Isembard IIMartin Barratt114R---7
MouseXDerek HallRRRRRx
MouseX2Derek HallRRRRRx
MicrobotJim ChidleyRRRR-x

As you can see, not terribly impressive. Except for Tim Foden’s FatCat. His was the only mouse that actually managed to perform a speed run and clocked up a perfectly respectable 11.6 seconds.

Only three mice completed a circuit run – a run around the outside squares of the maze. Fastest of these was Decimus 2B with a best time of 6.7 seconds. Next was FatCat with a best time of 7.4 seconds and last was HolyHead by David Hannaford with a time of 27 seconds.

The wall followers had rather better luck. At least none of them failed to find the middle:

FollowerOwnerTouchesTime
ErraticDavid Hannaford160
LeftyKen Hewitt164
McGuinnDuncan Louttit049
MusoBernard Grabowski036

Generally, problems with the maze and/or the lighting are something that the mouse builder must deal with but this, I think, may have been a challenge too great for meaningful competition.




Leave a Reply