Research on RBF neural network model compensation and adaptive control of robot manipulators

Abstract

According to the problem of accuracy of the manipulator's control trajectory, the adaptive control method based on model error compensation is studied. Because of the nonlinearity and uncertainty of the mechanical arm dynamics model, therefore, based on the computed torque control on PD, RBF neural network is used to approximate the modeling error and the uncertainty factor to achieve tracking control of robot arm. The Lyapunov stability of the system is proved. The adaptive law of network weights is established. Finally, the computer simulation results show that the method has good adaptive ability and tracking performance.