Internal-model control of vehicle chassis based on wavelet-network dynamic inversion method

Abstract

An internal-model control strategy of vehicle chassis system is proposed based on wavelet-network dynamic inversion method, to eliminate the coupling effect among the chassis subsystems. By analysing the chassis system reversibility, the paired relationship of the input and output variables is determined. Then, a wavelet-network dynamic inversion model of vehicle chassis system is obtained and connected in series with the original system. The chassis system is decoupled into three independent pseudo-linear systems, and decoupling between the control loops of vehicle chassis system is realised. An internal-model controller is designed to improve the system response. Further, a simulation and a driver-in-loop test using LabVIEW PXI and veDYNA are conducted to verify the control strategy efficacy. The results show that the proposed control strategy can eliminate the interference and coupling between the active front-wheel steering, direct yaw-moment control, and active suspension system, and improve the vehicle handling and status tracking stability.