| Active steering technology refers to the ability of the steering system to actively intervene when the car is about to fall into danger,thereby reducing the injury caused by the accident or avoiding the occurrence of the accident.This technology is one of the most important technologies in active safety technology,and it is also a necessary steering technology to realize autonomous driving technology.Therefore,the research of automobile safety technology based on active steering technology and autonomous driving technology has important practical significance.The control model of active steering technology involves uncertainty and nonlinear characteristics.Regarding the control goal,in addition to tracking the desired trajectory,the system also needs to meet the prescribe performance requirement,thus its control design faces great challenges.For the problem above,this thesis conducts research on the key technologies of active steering control.The major studies can be summarized as follows:(1)After considering nonlinear factors such as self-aligning torque,motor torque oscillation,gear backlash,etc.,a nonlinear model of the lateral motion of a two-degree-of-freedom vehicle is established.The desired trajectory of active four-wheel steering system is expressed in the form of holonomic and nonholonomic servo constraints.For the nominal system,based on the UdwadiaKalaba method,a constraint-following control is proposed to solve the problem of vehicle lateral trajectory tracking.(2)This thesis studies the trajectory tracking control of active steering system with prescribed performance requirements.In this thesis,a performance function inequality is constructed to limit the error variables.By state variable transformation,the bounded state variables are mapped into a new space,yielding new unbounded state variables.Correspondingly,a new transformed system and servo constraints can also be obtained.As a result,the control design for the new transformed system can drive the original system to strictly follow the prescribed performance and meet the original servo constraints.(3)This thesis studies the uncertainty problem in the four-wheel steering system,and proposes an adaptive robust constraint-following control.The uncertainty of the four-wheel steering system is fast time-varying,and the boundary information is unknown.In this thesis,an adaptive vector governed by a smoothzone adaptive law is employed to emulate the uncertainty bound.This estimated bound is used as the basis of robust control,so as to reduce the dependence of the system on the uncertainty information.Then,a robust constraint-following control method is designed to control the uncertain system and solve the trajectory tracking problem of vehicle lateral motion with uncertainties.(4)On the basis of the adaptive robust constraint-following control,this thesis proposes an optimization method of control parameters.The fuzzy set theory is introduced to describe the uncertainty information of the system,and the fuzzy dynamic model of the active steering system is obtained.Considering the performance and control cost of the active steering system,a performance index function is constructed.According to the fuzzy information of the active steering system and the defuzzification operation,the optimization problem of the control parameters is transformed into the optimal problem for the performance index function,and the optimal control parameter is obtained by solving the maximum value of the function.(5)Regarding the actuator—electric power steering system,a model-based compound control structure and a control parameter optimization method considering the change of driving torque gain are proposed.First,a composite control structure is proposed to enable the actuator to respond quickly to the control commands.Secondly,considering the influence of the change of the driving torque gain,a control optimization method is proposed to reduce the torque oscillation of the system. |
PDF Full Text Request