State observation of EV active suspension systems using event-triggered distributed moving horizon estimation

Abstract

This paper studies the state estimation issue of full-car active suspension systems of in-wheel motors-driven electric vehicles (EVs) under cloud monitoring. We propose a distributed moving horizon estimation (DMHE) method based on the event triggering mechanism (ETM) to deal with this issue. The full-car active suspension system is split into five interacting subsystems to lessen the calculation complexity of the DMHE. To improve the accuracy of the state estimation, the moving horizon estimation principle is used to handle the state and noise of the active suspension system. Then the packet loss event in the communication network is modeled as a random process that satisfies the Bernoulli property. Moreover, the boundedness of estimation errors of the proposed estimation method is proved under some sufficient conditions. The effectiveness of the algorithm is verified through the CarSim/Simulink joint simulation experiment and the half-car experimental platform.