Control of 3-DOF Robotic Manipulator by Neural Network Based Fuzzy-PD

摘要

A new control scheme of Back-Propagation Neural Network Based Fuzzy-PD has been proposed for control of a 3-DOF robotic manipulator in this paper. Combined with Back-Propagation Neural Network (BPNN), the performance of traditional Fuzzy-PD has been optimized by online calibration of the overlapping ratio of membership function. The overlapping ratio is firstly defined as the value of the intersection point between two adjacent membership functions. And then BPNN is utilized to optimize the overlapping ratio online. Simulations and real experiments with a 3-DOF robotic manipulator have been carried out to demonstrate the effectiveness and superiority of the proposed scheme.