Event-Triggered Adaptive Fault Tolerant Sliding Mode Control for FWID-EV

Abstract

In this paper, the problem of the Four-wheel independent drive electric vehicle (FWID-EV) driving on different roads with actuator faults is studied. Moreover, the issues of difficulty in observing the vehicle state, and external disturbances are researched, and the burden of shortage of Controller Area Network (CAN) bus communication resources is reduced. Firstly, the FWID-EV dynamic model with in-wheel motor faults and unknown disturbance is established. Different from the Unscented Kalman Filter (UKF) strategy, the novel extended states observer (ESO) is proposed to estimate both the vehicle states and disturbances to improve the controller accuracy and the ability of disturbance resistance of the vehicle. Then, the event-triggered (ET) mechanism is proposed to guarantee the saving of communication resources. Moreover, the adaptive fault tolerance sliding mode controller (AFTSMC) based on the ESO is proposed to improve the stability and safety of FWID-EV. The controller is made up of two layers. In the upper layer, the Adaptive sliding mode controller (ASMC) outputs the auxiliary longitudinal driving torque and yaw moments required for the vehicle to guarantee tracking performance. In the lower layer, different from the common optimal allocation strategy, the novel torque optimal allocation (TOA) strategy based on road friction coefficient and failure factors of in-wheel motors is proposed to output four in-wheel motors control input by optimizing both the utilization of the motor and tire under a smooth switching function. Finally, as the analysis of the result, the AFTSMC decreases the yaw error by 0.4 deg/s and lateral velocity by 0.15km/h in lane change maneuvers, and the allocation strategy improves a 10%-30% control margin in high friction road and the Hardware-In-Loop (HIL) experimental results validates the simulation results, which demonstrates the effectiveness of the proposed method through three illustrative cases.