Robust Lateral Motion Control of Electric Ground Vehicles with Random Network-induced Delays

Xiaoyuan Zhu,Zhang, H.,Junmin Wang,Zongde Fang

DOI: https://doi.org/10.1109/TVT.2014.2383402

2015-01-01

Abstract:This paper presents a lateral motion control strategy for four-wheel independently-actuated electric ground vehicles that use controller area network (CAN) as communication medium. The proposed controller design aims to guarantee the vehicle stability while tracking the desired yaw rate in spite of random network-induced delays that exist in both feedback and forward channels. By modeling the random network-induced delays in both channels as two homogenous Markov chains, statistic information of these delays are incorporated in the mode-dependent tracking controller design. The control law consists of the state-feedback control and the integral control. In order to fully compensate for the network-induced delays, a delay-free stochastic closed-loop system is first obtained in discrete-time framework by using a system augmentation technique. Then a robust linear quadratic regulator based H∞ controller is developed to achieve the tradeoff between the tracking error and the control input while also attenuating the effect of external disturbance. Considering the physical limitation of the in-wheel motors, the eigenvalue positions of the state matrix are constrained in a predefined area to further balance the control inputs and transient responses by using pole placement. Finally, an iterative linear matrix inequality algorithm is adopted to obtain the delay-dependent feedback control gains. Simulation results based on a high-fidelity, CarSim???, full-vehicle model show the effectiveness of the proposed lateral motion control approach.