Event-driven Tracking Control Of Assisted Motors Forelectric Power Steering Systems

Abstrct:In the past twenty years,electric power steering systems have beeninvestigated extensively. It is predicted that electric power steering systems willreplace traditional hydraulic power steering systems in the future. The thesisconsidersthe tracking control problem of electric power steering systems. The mainresults are summarized as follows:1) The thesis proposesan event-driven tracking control method for assited motors ofelectric power steering systems by setting a sample threshold. Whenever the changeof the output of the controlled system exceeds the threshold, the sensor transmits thecurrent output to the controller. Hence the sample is event-driven, not time-driven.We derive an event-driven PI controller. We show the validity of the PI controller byinvesitaging the stability and tracking performance. All the results are verified viasimulation.2) Based on physical laws, we constructed the dynamic models for all the componentsofanelectric power steering system. We then combined all these dynamic models tobuildthe model of the electric power steering system in Matlab. With this model, wefurther prove the correctness of the event-driven PI controller by simulation.Furthermore, in order to reduce the error of the output of the systemusing theevent-driven sample method, we designe an H∞robust controller.Simulation showsthat the H∞robust controller indeed can reduce the error.3) We constructe anexperimental platform for electric power steering systems. Ourwork includes choosing the hardware and software, combining all of them into anintergrated system, designing the communication protocols and implementing thePWM output. The experimental platform can be used to not only verify theevent-driven control method, but also develop and verify other control methods forelectric power steering systems.