Nonlinear robust control for electro-hydraulic servo systems with largely unknown model dynamics and disturbances

Abstract

In this paper, an adaptive sliding mode controller based on disturbance observers and neural network (NN)-based function approximators is introduced to improve the tracking performance of electro-hydraulic servo systems with largely unknown model dynamics. The RBF -based function approximators are employed to deal with unstructured uncertainties, whereas UDE-based disturbance observers are designed to estimate not only lumped mismatched disturbance but also matched disturbance. The derivatives of system states are obtained by using high-order Levant's exact differentiators. Finally, the adaptive robust control law is synthesized to attenuate the imperfections in disturbance estimation and NN-based approximation performances and guarantee high-accuracy tracking performance. The stability of the closed-loop system is analyzed by using Lyapunov theory. Comparative simulations based on an electro-hydraulic rotary are conducted using MATLAB/Simulink to verify the effectiveness of the proposed control approach.