An Improved Adaptive Neural Network Sliding Mode Method for Tracking Control

An improved adaptive neural network sliding mode control method is proposed for the trajectory tracking of the manipulator, considering the uncertainty of the mathematical model and the external disturbance. The dynamic model of the manipulator is divided into two parts: the nominal with the computational torque controller and the non-nominal part with the sliding mode controller, which was designed by the reaching law. For the designed sliding mode controller, radial basis function (RBF) neural network is used to compensate the uncertainties. At the same time, the symbol function is replaced by the saturation function to further suppress system chattering. The global stability of the system is proved by Lyapunov function. The simulation results of two-link manipulator show that the proposed method can effectively overcome the uncertainty of the system and track the desired trajectory with fast convergence and small error.