Robust Tracking Control of Flexible Joint with Nonlinear Friction and Uncertainties Using Wavelet Neural Networks

A new combination method for the robust high precision position control of flexible joint with nonlinear friction and uncertainties compensation is proposed in this paper. The global control system is designed based on backstepping like technique in order to suppress the flexibility and disturbance, and the wavelet neural networks approach is utilized especially to estimate the nonlinear terms including friction and uncertainties locally without predicting the upper bound. From the Lyapunov stability analysis, adaptive laws which are used to update the network weights are induced. The stability of the overall closed loop system is proved to be UUB stable. Finally, the computer simulation results show the good position tracking performance and robustness of the proposed control strategy.