NZSM Online

Get TurboNote+ desktop sticky notes

Interclue makes your browsing smarter, faster, more informative

SciTech Daily Review

Webcentre Ltd: Web solutions, Smart software, Quality graphics

Spotlight

Micromouse

The "micromouse" is a fully autonomous, intelligent, mobile electronic device. The concept was first proposed about 20 years ago, and a competition was designed to test which device was the best. It seemed the best way to do this was to place the device in a maze (16 x 16 squares), and have races to see which gets to the end in the shortest time, hence the name "micromouse".

The competition has evolved somewhat, with the rules now being that the mouse can spend a total of 15 minutes in the maze, a certain length of time of which will be spent mapping the maze. At any time the mouse can go back to the start of the maze, and go for a "speed run", where it attempts to get to the middle in the shortest possible time. Note that the fastest route may not necessarily be the shortest one, as the mouse can travel considerably faster down straights than around corners.

From reasonably humble starts, the micromouse competition is now an international event and various institutions around the world devote enormous resources into it. In Hong Kong, Japanese and Singapore universities it is not uncommon to have entire labs -- and in some cases a whole building -- complete with dozens of support staff devoted to research into the micromouse.

In New Zealand, the contest has been run since 1990. In 1993, Gary Brightwell's mouse Penfold (named from the DangerMouse cartoon) became the first New Zealand entrant to ever successfully solve a maze in competition. When we took Penfold to Australia for their national competition, we were let down by mechanical problems. This is becoming our biggest problem: the precision engineering of the gears, chassis and so on, that is required for competitive "mousing".

Some Australian universities provide their students with kitsets, where the amount of student effort to get the thing running is minimal. We disagree with this approach, and it takes most of the learning aspect out of the project. However, we operate on a minute budget, we have to scavenge gears from clocks, sensors from old computer disk-drives, a real "do it yourself, Kiwi ingenuity" approach. The students also learn a tremendous amount about mechanical, electrical and software engineering, as well as sensor technology, systems control and, to an extent, even theoretical physics. All these skills are in high demand in the electronics industry, so the whole exercise has a very real value.

As for last year's competition in Auckland, it was not all smooth sailing...

We arrive to find that Penfold has not survived the trip from Hamilton, and is more like a micro-mule than anything else, and stubbornly refuses to move. That's okay, we've got replacement parts for everything, except the sensors, but that's not likely to have failed. WRONG!

It appears that one of the infrared sensors that Penfold uses to find the walls of the maze has failed. So at 10:30 with the competition due to start at 3:30, Gary jumps in his car, and takes Penfold on a whirlwind trip back to our lab in Hamilton, leaving me to fret in Auckland. My nerves aren't holding out too well, so it is with considerable relief that I see Gary and our technician Alastair come through the doors at 3:10. Okay everything's fine now. WRONG!

In their haste, they forget to bring back the download cable which transfers the program into the mouse's memory. We are totally unable to scavenge anything, so it's into the car for a hurried trip to Dick Smith, a considerable distance away. Start time arrives and still no sign of them. I instruct the organiser to proceed without us, and the competition starts. Just as the first device is placed in the maze, through the door bursts the recovery team, so now surely everything must be okay. WRONG!

They have all the pieces, but it still needs to be put together. Out comes the veroboard, a few ICs, I have students stripping wire all over the place with their teeth, spitting out bits of plastic all over the show, and Alastair doing power-soldering like a man possessed. With 30 seconds before our run is due to start, Alastair announces "I think that's it!" Well we can forget any pretense at composure as it appears that the cable (with drivers) actually works, but there's still no time to test it, and into the maze goes Penfold.

It gets off to a good start, sets a New Zealand record with the distance it reaches and, then, with one square to go until the middle, it stops. It just stops! Not only could you hear a pin drop, I'm sure you could hear it bounce! Finally, after about 30 of the longest seconds of our lives, Penfold decides that it has had enough of baiting us, turns the corner into the middle of the maze, and becomes the first New Zealand-made mouse to ever complete the circuit.

I'm afraid we had to party that night to celebrate.

Dale Carnegie

Dale Carnegie is director of Waikato University's Microelectronics Group.