Behavior generation of bipedal robot using central pattern generator(CPG) (1st report: CPG parameters searching method by genetic algorithm)

Abstract

In biological systems, the rhythm generator mechanism called Central Pattern Generator (CPG) has been involved in rhythmic activities, such as locomotion, respiration, heart beat, etc, and various mathematical neuron models are proposed to represent its oscillation signals. In this paper, the CPG network which consists of the Matsuoka neuron model are introduced to realize the locomotion of a bipedal robot. The outputs of neurons are utilized as the target angles of corresponding joints. A parameters searching method using Genetic Algorithm, which has several steps to find out large number of parameters according to structure of CPG network, is proposed to obtain the CPG parameters and realize the CPG network efficiently.