14. SUGGESTED FIRST MICE
I have seen many mice which worked and many more which did not. I recommend that the ideal first micromouse is a wheel-chair style with two stepper motors and IR sensing looking down on the walls, because:
- There is no need for wheel motion sensors, so
- The simplest possible electronics,
- Simple mechanical arrangement, and from these points,
Figure 2 is a sketch of such a device. The PCB is shown in outline, and is about 55 mm above the floor, the motors shown are 40x40x30, but any motor of about that size will provide enough torque to drive a micromouse. The more steps per revolution the better, in general. The number of cells used can be found by trial and failure, more is not necessarily better - high voltage on steppers means high current consumption which means shorter charge life.
The wheels may be driven by gearing or by using a pulley on the stepper shaft using an o-ring as both drive-belt and tyre. The torque characteristics of economical steppers make it necessary to use reduction gearing or a very small output wheel. The o-ring belt/tyre method is the simplest form of this, that fits the geometry, but if you can make a gearing system, do so.
Keep the chassis narrow to allow latitude in cornering without hitting walls, and to make it easier to get out of dead ends.
Juggle the centre of gravity by moving cells to make the micromouse balance on the wheels. Sliders may be made from any 'slippery' material. Mount one or both on springs to allow for the inevitable small bumps in the micromouse maze surface. The obvious alternative of small casters may be unreliable because of dirt and the small size.
Most builders probably prefer to buy a ready-made PCB. A micromouse can be made using a wire-wrapped circuit board, which allows full rein to the designer's ingenuity. If you prefer to buy, a partial kit, consisting of a PCB, circuit and construction diagrams, and building and operating instructions will be available in the future. This design will allow for code to be downloaded from a support machine over a simple serial interface, to dispense with the need for an EPROM blower. Later, an almost complete kit of parts is to be made available for purchase to allow anyone to make a micromouse which will work with the minimum of trial and error. The micromouse might not win the 'professionals' class, but should reach the centre of the micromouse maze and can be modified by a determined team to improve the performance as much as their expertise will allow.
Figure 2: Example of a Micromouse
This article was written by Alan Dibley and was combined with a list of references provided by me, Martin Smith. We produced the competition guide to help future micromouse builders. The guide was distributed at cost by the IEE in 1993. As the competition guide is no longer available I have scanned an old text copy and read it with an optical character reader. I have omitted information that only applied to 1993 and might mislead current builders. Any errors or other omissions are my responsibility. Please contact me if you notice any at: Professor Martin Smith