Locomotion of a quadruped robot using CPG

Abstract

It is well known that the rhythm generator mechanism called central pattern generator (CPG) controls rhythmic activities, such as locomotion, respiration, heart beat, etc. in biological systems, and various neuron models are proposed. The CPG has attractive features such that (i) it generates periodical signals persistently, (ii) return to the original oscillation if disturbances are removed, (iii) the mathematical conditions for oscillation are proved, and so on. A CPG network, which consists of the Matsuoka model neurons, is introduced to realize the locomotion of a quadruped walking robot and the response to disturbances is discussed. The outputs of neurons are utilized as the target angles of corresponding joints and the efficiency of the proposed method is examined through experiments with a quadruped walking robot.