An adaptive neural network sliding mode controller for nonlinear active suspension with uncertain parameters and non-ideal actuators

Abstract

Active suspension systems (ASSs) are contributed to improving ride comfort and maneuverability. However, practical ASSs commonly suffer from nonlinear characteristics, uncertain parameters and non-ideal actuators, which always significantly deteriorate the control performance in practice. To overcome these issues, this paper proposes an adaptive neural network sliding mode control (ANNSMC) strategy for an ASS to achieve suspension performance improvements. Firstly, the skyhook system as a reference model which doesn’t require real-time measurement of road input is adopted to provide reference trajectories for the sprung mass displacement and velocity. In addition, an adaptive radial basis function neural network is designed and presented to deal with the effects of uncertain nonlinear functions in the dynamic system. Furthermore, the stability of the closed-loop system is proved by the Lyapunov stability theory. Finally, the effectiveness of the proposed controller is verified by numerical simulation and experimental platform, and the advantages of the proposed ANNSMC controller in suppressing suspension vibration are illustrated by comparing the performance of passive suspension (PS), sliding mode control (SMC) and linear quadratic regulator (LQR) controlled active suspension under different road excitations. The simulation and experimental results further demonstrate that the proposed controller can effectively suppress sprung mass vibrations and offers superior control performance despite the existence of nonlinear dynamics, uncertain parameters and non-ideal actuators.