Research On Steer-by-Wire Vehicle Control System Based On Sliding Mode Control Theory And The Lumped Uncertainty Estimation With Extreme Learning Machine

With the development of automobile industry,people always have a higher standard of automobile comfort and safety performance.The steering system is one of the key technologies of automobile chassis design,which plays a significant role in improving driving comfort performance and stability.In recent years,the vehicle with steer-bysystem is becoming more and more popular among the scholars,automobile manufacturers and electronic control device suppliers due to its merits of high level of intelligence,saving vehicle interior space,meeting the requirements of lightweight,high security,strong comfort performance.Based on the research of sliding mode theory,this dissertation studies and discusses the control strategy design of steer-by-wire system.Firstly,the operation principle of the steer-by-wire system is analyzed detailedly in this work for identifying the function of every steering control component,the dynamic equation of each component is constructed,and then the synthetic model of the system is obtained through comprehensive analysis,which provides accurate mathematical model for the control strategy design of steer-by-wire system.Then,in this dissertation,a novel extreme-learning-machine(ELM)-based robust adaptive integral terminal sliding mode(AITSM)control strategy is developed for the precise tracking control of a steer-by-wire(SBW)system with uncertain dynamics.Different from conventional ELM using least square optimization approach,the ELM in this work is designed to adaptively estimate the lumped uncertainty from the perspective of global stability of the closed-loop system.The proposed control not only ensures the finite-time error convergence but also effectively estimates the lumped uncertainty via a single-hidden layer feedforward network(SLFN)with ELM.At the same time,traditional linear control,sliding mode control and nonsingular fast terminal sliding mode(NFTSM)control are introduced as simulation comparisons,which proves the superiority of the proposed control strategy.In addition,considering the discretization properties of the most digital control systems,this paper also proposes the discrete integral terminal sliding mode control for the steer-by-wire system,and briefly introduces the discretization method of the steer-by-wire system model,the design method of the discrete control strategy and the stability proof method.Finally,according to the characteristics of vehicle steering process,three groups of continuous simulations and two groups of discrete simulations are designed to verify the effectiveness and robustness of the proposed controllers.The simulation results illustrate that extreme learning machine based robust adaptive integral terminal sliding mode control can not only effectively estimate and compensate the lumped uncertainty,improving tracking accuracy of the system and becoming insensitive for external uncertainty,but also can eliminate the reaching phase of the sliding mode and improve the convergence rate of the closed-loop error dynamics by the proper selection of the initial value in sliding surface.In the discrete control strategy,the simulation results also verify that the discrete integral terminal sliding mode control strategy has strong robustness and tracking accuracy.