CPG control model of snake-like robot parameters of optimization based on GA

Abstract

In this paper, we present a biomimetic architecture to solve the difficulty in control of a snake-like robot with large degrees of freedom based on the structure of biological snakes and snake-like robots and their rhythm characteristic. Based on the Hodgkin-Huxley's CPG (central pattern generator) model, we established the CPG-networks to control the snake-like robot's gait. Compared with the previous research, parameter optimization is carried out for two typical gait patterns by means of genetic algorithm. It is proved that the CPG model can realize control of the snake-like robot gait by the simulation results. Finally, the aspects of future researches are discussed.