Neural network adaptive control of nonlinear systems preceded by hysteresis

Abstract

Neural network adaptive control is proposed for a class of nonlinear system preceded by hysteresis. A novel model is developed to represent the hysteresis characteristics in explicit form. Furthermore, the auxiliary variable of the proposed model is proved to be bounded, which is essential for controller design. Then, neural network adaptive controller is directly applied to mitigate the influence of the hysteresis without constructing the hysteresis inverse. The updated law and control law of the controllers are derived from Lyapunov stability theorem, so that the boundedness of the close-loop system is guaranteed. Finally, the experimental tests are carried out to validate the effectiveness of the proposed approach.