Adaptive Neural Networks Control for Half-Car Active Suspension Systems with Prescribed Performance and Actuator Fault

Abstract

This study proposes an adaptive neural backstepping control scheme for a half-car vehicle suspension system considering the displacement constraint of chassis and actuator failures. The unknown functions caused by the different passenger masses and uncertain factors are estimated by neural networks. To guarantee the chassis displacement within limited constraints, the prescribed performance function is used to describe the convergence rate of tracking error and ensure the maximum overshoot within the boundaries. The vertical displacement and pitch angle of the half-car model are considered simultaneously to improve the riding comfortability and handling stability of suspension performance. The comparative simulation examples will be realized to show the feasibility and effectiveness of the developed method.