Robot Arm Trajectory Tracking based on adaptive neural Control

Abstract

Control system of robot arm is a multi-variable, strong coupling, and highly nonlinear uncertain system, and trajectory tracking requires the robot manipulator to move according to a desired trajectory that has been given. According to friction and disturbance problem, an adaptive neural network control is proposed. Neural network is used to compensate the dynamic uncertainty of system, and the neural network approximation error and friction and disturbance part are compensated by an parameter adaptive compensation. The simulation results show that this algorithm can improve the effectiveness and accuracy of mechanical arm trajectory tracking.