Adaptative motion control system of a snake-like robot using a neural oscillator netowork

This paper presents a design method for a phase lag adjustment system using neural oscillators for a snake-like robot. The neural oscillator or central pattern generator (CPG), one of biomimetic behavior control system has attract attentions as we can see the similar rhythmical system in the natural creatures like fish swimming, bird flapping, etc, and these behaviors are considered to be optimized to their working environments. We found that the neural oscillator group had the possibility to adjust the robot behavior into working environment by using physical feedback signals from the working environment. The neural oscillator network adjusts their phase lag between neural oscillators to adapt their motor torques and trajectory to the working environment. In this paper, we present an phase lag auto-adjustment system using neural oscillator and its application into motion control of a snake-like robot, and the efficiency is evaluated through simulations and experiments using a 8-links snake robot.