Adaptive Multiple-surface Sliding Control of Hydraulic Active Suspension Systems Based on the Function Approximation Technique

Abstract

In this paper we propose an adaptive multiple-surface sliding controller (AMSSC) to control a non-autonomous quarter-car suspension system with hydraulic actuator. Due to the spring nonlinearities, the system property becomes asymmetric under the system’s own weight. Besides, because precise parameters of practical systems are hard to obtain, the system uncertainties should be dealt with. In this paper, these uncertainties are assumed to be lumped into three unknown functions such that the system model has both matched and mismatched uncertainties. Because the bounds of some of time-varying uncertainties are unavailable, traditional adaptive schemes or robust strategies are infeasible. To deal with this problem, a function approximation based adaptive multiple-surface sliding controller (AMSSC) is proposed in this paper. The multiple-surface sliding controller (MSSC) is used to cope with mismatched uncertainties while the function approximation technique is used to represent those uncertainties as finite combinations of basis functions. Adaptive laws for the approximating series can thus be derived based on the Lyapunov-like approach to ensure the closed-loop stability. Convergent performance of tracking errors can be obtained to improve the ride quality. Because the state measurements of the unsprung mass are lumped into the uncertainties, there is no need to feed back these signals with the proposed method. Therefore, the hardware structure can be simplified in the actual implementation. Computer simulations are performed to verify the effectiveness of the proposed strategy.