| When people drive a motor vehicle, the road traffic safety accidents happen frequently caused byindividual factors. The driver’s behavior is a major reason which induces traffic accidents. So it hasbecome the focus on how to analysis and research on the characteristics of driver’s driving behavior inorder to build a safe and comfortable driving environment. And given of defects of the high test cost,long time and influenced by environmental conditions, in this case, it is a more accurate andreasonable for us to develop driver model which can reflects characteristics of human driver. It isadvantageous to the vehicle experiment validation, analysis and aided design of vehicle.Focusing on the steering control model of driver, this paper conducts in-depth discussions on theLinear Quadratic Regulator (LQR) method of driver’s steering control model, Neuromuscular System(NMS) dynamic model and vehicle stability control model. Concrete contents are summarized asfollows：(1) Firstly, based on LQR and given the characteristic of driver’s time delay, a driver model isdesigned. Focusing on the driver’s steering control, discrete “optimal-preview” of vehicle-road isestablished. The designing model meets the vehicle dynamics and the optimal control theory, andadopts the multi-point prediction aiming mechanism. According to different weight value, simulateand analysis the performance of the trajectory followed. According to the feedback coefficient matrix,analysis internal driving characteristics of optimal driver. Considering the characteristic of driver’stime delay, takes the form of state shift register to set up the model. The simulation results withdouble shift line test method to verify the effectiveness of the driver model.(2) Secondly, a mathematical model of driver steering control including neuromuscular systemdynamic driver is proposed. According to the specific process and the mechanism of the muscle’sreaction, NMS model is developed which contain reflection controller, active stiffness and the internalreference model. According to the dynamic coupling of the arms and steering system, dynamicequation is deduced that is the dynamic equations of arm and steering system. The transfer function ofthe coupling system is obtained by the method of system identification, so it is better to study thedriver’s learning mechanism, and verify that the accuracy of the internal reference model influence thedriver’s steering control. And Given the influence of steering torque feedback, neuromusculardynamic parameters are verified with simulation for the influence of the trajectory trackingperformance. (3) Finally, based on the mathematical model of driver steering control including NMSdynamics, the vehicle stability control system is presented. Given the nonlinear characteristics of tire,vehicle dynamics model is established based on vehicle dynamics and the magic formula. Thenaccording to linear dynamic model of the vehicle and the optimal control theory, a kind of additionalyaw moment control method is designed, so as to improve vehicle stability. And the steering stabilitycontrol algorithm is validated to guarantee the stability of the vehicle in different adhesion coefficientroad.To summarize, on the one hand driver steering control model is designed in this paper. Troughthe double line simulation, trajectory tracking performance is verified under different control value toselect optimal driver steering control model. On the other hand, the fusion of NMS dynamic model tothe driver steering control model, the results indicate that the driver’s NMS dynamic parameters(reflection gain, stiffness factor) and the accuracy of the internal reference model, have importanteffect on vehicle trajectory tracking. On this basis, the vehicle stability control model is established,not only improves the stability of the vehicle on different adhesion coefficient of road surface, and theresults in related fields have important theoretical significance and engineering application value. |
PDF Full Text Request