EXPERIMENTAL INVESTIGATION OF AN ADAPTIVE NEURO-FUZZY CONTROL SCHEME FOR INDUSTRIAL ROBOTS

Abstract

This paper presents the application of an adaptive fuzzy logic controller with feed-forwardcomponent (AFLCF) to the Selective Compliance Assembly Robot Arm (SCARA Robot). The feedforward torque component is computed on-line using an artificial neural network (ANN) which hasbeen trained off-line. This feed-forward component is designed to deliver the ideal torquecomponent to the robot derivers. The feedback fuzzy logic control (FLC) component is made tokeep the stability of the closed loop system. As the FLC is dependent in its rule base, here, acompact rule base is used. It consists of only four rules per each degree of freedom (DOF). The FLCensures closed loop stability in the sense of Lyapunov and is valid for second order nonlinearsystems. Furthermore, adaptability of the FLC has been achieved to enhance the trackingperformance. The theoretical background of this control algorithm has been published in[1].Using SCARA robot as the testing platform, here, experimental results are presented for thefollowing five controllers: the conventional PD controller, PD controller tuned by fuzzy system(PDT), the FLC, Adaptive FLC (AFLC), and finally the AFLCF. The controllers are testedexperimentally at the same initial conditions to make fair comparison between their performances.Results show that the investigated AFLCF has outperformed the other controllers.