Micromouse Guide


A micromouse is a device with:

  1. A chassis
  2. A drive motor or motors to move it
  3. A steering and turning method
  4. Sensors to detect the presence or absence of micromouse maze walls
  5. Control logic (a computer and peripheral circuits) to oversee the action of the rest and solve the micromouse maze
  6. Batteries to provide power
  7. Designer(s) and builder(s).

Because the cells of a micromouse maze are based on an 18cm square grid, mice are typically about 12 to 15cm across at the bottom, but may overhang the 5cm high walls. The maximum horizontal dimension is 25cm square. This allows the use of downward-pointing infra-red sensors to detect walls (if you want to do it that way), and means that the computer board does not have to be sub-miniature. The author used Sinclair ZX80's or ZX81's in five mice of the Thezeus family and used a purpose-built processor board only for number six. (Note: All my mice, and most other successful mice, have been male. I don't know why and offer no comment. There was at least one exception, called Maisie, so it is not a hard and fast rule).

This article offers a range of ideas to get your mind churning with the thought 'I could do that' or 'We could make a Micromouse'.

I have built six mice which competed in the UK, European and world competitions. All were built using easily available technology, none cost more than about £100 for components (many of the bits were scrounged or found in electronic junk stores or Sunday markets) and all achieved success in some way. Here are some ideas offered as being possible routes to success. First; some rules and disciplines. Sorry if they seem like a lecture, but you will learn the hard way otherwise.

Keep it simple, because if you don't, it will not work. 'Simple' means that you must understand the technology you use. Example: will you use stepper motors or DC motors to drive the device. If you have used stepper motors before use them again - you do not want to have to learn about the control of DC motors and the use of feed-back loops on top of the other challenges. And you will have challenges - or disasters.

Take one step at a time, and get it right. You do not need to build a complete working micromouse at the first attempt. Start by making a rolling chassis with a motor or motors on it. When you can make it move forwards and steer or turn by sending signals down a cable from a fixed computer or from a simple switch box, start adding other elements. Experience with devices such as Turtles as used in some school projects is useful.


The first micromouse you design might be a disaster, so it is a good idea to make it fairly crude. Then you will not feel too inhibited about dismantling it when you discover the shortcomings. A beautifully crafted device with a pretty finish will not get to the centre if it does not work correctly. That does not mean it can be a rats-nest of loose cables and components trailing on bits of twisted-together lash-up wire, but consider the possibility of making a temporary chassis of wood. It can be drilled, cut, glued, screwed and sanded with hand tools, and be light yet strong enough. When it works, replace it with a smarter one.

The main circuit board may be part of the chassis. A piece of epoxy PCB is strong enough to support a set of IR diodes or micro-switches, or even be the mount for motors and spindles, with no sub-structure. Make wheels from wood or plastic, turned on a small lathe - a Unimat or similar machine, and use rubber sealing "0"-rings as tyres, glued with cyano adhesive or used as described below, as drive-belts. Keep the weight down but do not make a fetish of it. The batteries and motors are heavy anyway and saving a tiny amount of weight here and there is a waste of effort, but don't build it like a tank. It will not topple at the speed it will achieve on your first effort, so do not worry if it looks a bit top-heavy. Micromouse mazes used in competition are all fairly flat and smooth. There is no limit on the permitted height of a micromouse.


There have been mice imitating almost every shape of full-sized wheeled transport and a few new ideas. It is difficult to cover the full range of possibilities, but here are a few:

1. The wheel-chair.
This arrangement uses two independent driving wheels, one each side of centre, and rollers or sliders at front/rear to stop it tilting. Steering is achieved by varying the speed of the two motors. Commonly uses stepper motors.

2. The tricycle.
Like a child's tricycle, one steering wheel at the front, two wheels at the back. The tricycle may be driven by one motor through the front wheel, or by two motors through the back wheels.

3. Square.
The most famous square machine was "Thumper" who stormed the European scene in the early '80s. The chassis just fits in the 16.8cm space between walls and rests lightly against them with rubber rollers. Runs on four wheels mounted on turntables. The wheels can be revolved on their turntables to face north-south or east-west. At corners, it stops, revolves the wheels to point in the appropriate direction, and restarts. It is slow at corners but reliable steering on straights allows it to run very fast along corridors which it has already explored. Unbeatable in its time, and it talked, using one of the first "speech chips" but it was mechanically complex.

4. Lots of others.
I remember a beautiful piece of engineering, which rolled along the top of the walls. My first micromouse was a tricycle with two front steering wheels and one rear driving wheel. It stopped at corners and turned by lifting itself up on a central foot, on which the whole structure revolved 90 or 180 degrees and was let down to the ground again. It was a very complicated mechanism, but seemed a good idea at the time, and worked reliably but very slowly. There are even bicycle mice.

Beginners should stick to either 1 or 2.

Note the strategies available for steering and turning. Steering is keeping the micromouse reasonably central in the micromouse maze corridors. This is not as easy as it first appears, even at low speed, and is even more difficult at high speed. Consider mechanical steering, rather than electro-mechanical steering, only if you have someone in the team with excellent mechanical aptitude. It is usually necessary for any micromouse of type 1 or 2 to turn in its own length, to get out of dead-ends, but there are double-ended mice. A wheel-chair turns on the spot by reversing one motor, but a tricycle is usually able to turn the front wheel through 90 degrees, and then either use the single driven front wheel to turn on the spot, or drive the two rear-wheel motors in opposite directions. Driving through the front wheel is simpler, but the motor must be mounted on the steering assembly or use a clever transmission method. Stopping at corners and revolving on the spots may be more reliable than taking corners by driving round a 90 degree turn, because the steering system has little or no input from sensors during the turn-but there is a time penalty.

Next page>>


Please note that TEE is not responsible for the content of external internet sites that may be linked to from this site.