Radial basis function neural network based adaptive fast nonsingular terminal sliding mode controller for piezo positioning stage

Abstract

This paper presents an adaptive fast nonsingular terminal sliding mode control base on a neural network based approximation technique to control the position of a piezo positioning stage (PSS). The proposed terminal sliding mode control can provide faster convergence and higher precision control while maintain its robustness to uncertainties. In the proposed control scheme, the combination of the fast-nonsingular terminal sliding mode control and neural network, which can precisely estimate the uncertainties in dynamic of the PSS system by employing an online tuning scheme, is a promising control approach for actuator systems. In addition, the robust control term is adopted to compensate the modeling error and ensure the robustness corresponding to a bounded disturbance. Stability of the closed loop system is analyzed and proved by using special Lyapunov functions. Experiment results strongly confirm the effectiveness of the proposed control method.