Fuzzy path tracking control of a 5-DOF experimental robot

Abstract

In this paper, a fuzzy controller is proposed for a five degrees of freedom robot and implemented to an experimental robot to track the desired paths. The desired paths are constructed by a trajectory planning unit. This unit generates the position and velocity of each motor axis for the controller as its reference signals. For a given path, the proposed fuzzy system, utilizes the errors between the actual position and speed with the reference signals, and produces the necessary control actions to the robot motors based on fuzzy control strategy. The fuzzy rule base which is formed based on experimental results, defines the motor torque for each axis correspond to relevant error and rate of error. A circular path is designed to show the effectiveness of the proposed method. The presented results emphasize that a satisfactory tracking precision is achieved. The tracking errors are also compared with a conventional PID controller.