ROBUST CONTROL BASED ON NEURAL NETWORKS AND RISE SATURATION CONTROLLER FOR ACTIVE MACPHERSON SUSPENSION SYSTEM TO IMPROVE RIDE COMFORT

Abstract

The paper studies on the control method based on the neural networks and saturation RISE controller for the active Macpherson suspension to improve the ride comfort, whereas, the suspension dynamics include the nonlinear uncertainties and exogenous disturbances. Feedforward neural network with the universal approximation capability is exploited to compensate for the nonlinear uncertainties and the robust saturation RISE controller with the control force limited in a priori limit is used to robustly regulate the vertical displacement of the sprung mass to improve the ride comfort. The Matlab simulations are performed to show the effectiveness of the proposed method in both time domain and frequency domain in comparison with the active suspension with PID controller, the semi-active suspension with a modified Skyhook control and the passive suspension.