This robot uses eight servo motors to control its four legs. The hardware used is an Arduino Mega that has enough PWM pins to control the motors. Nonetheless, it uses ultrasound and accelerometer sensors to be used in the fitness calculation of the genetic algorithm (GA).
About the genes:
The genes are formed by a vector each, that stores motor's movements for each leg. The first genes generation are randomly created, filling up all the vectors (those vectors have a fixed size).
The crossover method:
The half best genes suffer crossover with 1% mutation. And the other half simply suffer crossover.
The fitness function:
The Incredible Beast uses ultrasound sensor to get the distance from a reference. This distance value is one of the values used for the fitness value of the genes.
Secondly, the robot also has an accelerometer. So, it is possible to get the 3-axis values and determine whether the robot is rightly oriented or not.