Terminal sliding mode robust control for manipulator based on adaptive neural network

Aiming at the uncertain manipulator system with unknown load moment and model error, considering the dynamic characteristics of the motor, a terminal sliding mode robust control method based on adaptive neural network is proposed. Firstly, the model of manipulator system including uncertainty and motor dynamics is established. Then, the chattering phenomenon of traditional sliding surface is suppressed by designing terminal sliding surface, and the terminal sliding robust control law is proposed. Meanwhile, the RBF neural network is introduced to accurately estimate the uncertainty, and the adaptive law is designed to update the weight vector in real time, so as to effectively improve the control effect. The simulation results show that the proposed method can quickly estimate the uncertainty and accurately track the command signal. The tracking errors of angle, angular velocity and angular acceleration are 0.1°, 0.1°/s and 0.1°/s² respectively. The uncertain estimation error is only 0.2 A, which realizes the high-precision robust control of the manipulator system.