Adaptive Optimal Locomotion of Snake Robot Based on CPG-Network Using Fuzzy Logic Tuner

Abstract

Periodic locomotion of animal bodies with large degrees of freedom is known to be realized by network of central pattern generators (CPGs) that are distributed in spinal cord (in vertebrates) or nerve cords (in invertebrates). In this paper, optimization of a controller for a snake robot locomotion based on CPG-network is presented. CPGs are modeled as nonlinear oscillators for each joint. The inter-joint coordination is achieved by altering the connection weights between joints. Genetic algorithm (GA) is used to optimize CPG parameters and connection weights in terms of moving speed. We proposed a new method that can be used as on line detection of changes in environmental conditions. Effect of friction coefficients on optimal parameters is next investigated. Results are utilized to design a fuzzy logic tuner with the goal of maintaining optimality of the locomotion while snake robot moves in different environmental condition (surfaces with different friction coefficients). Optimal CPG-network parameters are also obtained for snake robot with different numbers of links. Results indicate that the fuzzy rules can be expanded for snake robot with any numbers of links. This paper is a step towards designing an optimal CPG controller with improved environmental adaptability.