Evolving Leg Cycles To Produce Hexapod Gaits

Abstract

Gait generation for hexapod robots is accomplished by dividing the problem into two parts: leg cycle learning and gait cycle learning. Servo pulses required to generate a single leg cycle are learned by a cyclic genetic algorithm. This learning takes into account the peculiarities of the leg's capabilities plus determines the proper sequence of pulses needed to generate smooth movement by the servos. The best means of combining these leg cycles into a gait cycle is learned by a genetic algorithm. This part requires the learning system to choose the best leg cycles for each leg and to coordinate their movement. In this paper, we describe an application of this method to learn gaits for a simulated hexapod robot.