New neural networks for adaptive control of robot manipulators

Abstract

The use of neural network for controlling of a robot is discussed in this paper. A new neural network is proposed to model and control a robot. The proposed network is a modification of the original Elman network. Also, a new control system design based on the neural network model of the robot dynamics for trajectory tracking control is studied. The proposed control system consists of a neural controller, a neural model of the robot and a P-controller. Simulation results show the effectiveness of using the neural network model in the design of the robot controller.