Feedback error learning control using adaptive fuzzy network to control one linear actuator hopping robot

Abstract

This paper addresses the control method of one linear actuator hopping robot in a plane by using feedback error learning scheme. Adaptive fuzzy network was used in this scheme, and backpropagation techniques utilized as learning algorithm. The robot consists of a body and a leg, which are in contact with a sufficiently wide horizontal ground; both are fixed rigidly. We consider utilizing the starting time of actuation of actuator - when the robot is in stance phase - to control the gait. By simulation, continuous hopping gait was realized.