Robust Control Of Electric Vehicle Lateral System Under The Influence Of Communication Network

In recent years,the research on the stability and maneuverability of electric vehicle lateral dynamics has been widely studied by many scholars.Electric vehicle may skid or even instable when turning at high speed,changing lanes or low road adhesion coefficient.Therefore,how to improve the lateral stability and maneuverability of electric vehicle is an urgent problem to be solved.As the development of electric vehicle and intelligent vehicle span many fields,such as electronics,machinery,information,automation and so on,the control design of electric vehicle lateral dynamics is a very complex problem.On the other hand,the development of network communication can improve the overall performance of electric vehicle due to it’s sharing information resources,reducing system wiring,easy to expand and maintain,and increasing the flexibility and reliability of the system.Meanwhile,it also brings some problems,i.e.quantization error,data packet loss,etc.,which will decrease the performance of electric vehicle.Therefore,it is of practical significance and value to study the robust control of electric vehicle lateral dynamics with the influence of network constrains.The main works of this thesis are given as follows:1)Research on the robust control problem of electric vehicle lateral dynamics under ideal network communication.Firstly,the tire cornering stiffness,controller and actuator are compensated via norm bounded uncertainty and randomly occurred uncertainty,respectively;secondly,in order to ensure that the sideslip angle and yaw rate satisfy H_∞ and L₂-L_∞ performance respectively,a new cost function is constructed to simplify the performance analysis condition;Finally,in order to obtain the required non-fragile controller,we use two-step method to decouple the uncertainty problem.Finally,the control design condition is given in forms of linear matrix inequalities.2)Research on the filtering problem of electric vehicle lateral dynamics with the influence of multi-channel sensor failure,quantization and data dropouts.Firstly,we use T-S fuzzy modeling technology to solve the nonlinear challenges of the mass and moment of inertia of electric vehicle lateral dynamics;Secondly,considering the quantization and data dropouts in both the measurement and performance outputs,we obtain the T-S fuzzy lateral filtering error system of electric vehicle with uncertainty;Finally,by using Lyapunov stability theory and linear matrix inequality（LMI）technology,the one-step design condition of the filter and two dynamic quantizers is given,which ensure that the electric vehicle lateral filtering error system is stochastic stable with prescribed H_∞performance.3)Research on the output feedback control problem of electric vehicle lateral dynamics based on state reconstruction with the influence of quantization and data dropouts.In order to ensure that the sideslip angle and yaw rate satisfy H_∞ and L₂-L_∞ performance respectively,we construct a new Lyapunov function,and give the one-step design condition of quantizer,observer and controller in forms of linear matrix inequalities.Finally,the single-lane change maneuver on electric vehicle is carried out by the joint simulation of Car Sim and MATLAB/Simulink with different data dropouts rates to validate the effectiveness of the proposed method.