Study On Vehicle Lateral Assistant Power Control Based On Driver Operating Characteristics

The vehicle lateral assistant system as an important part of the vehicle active safety control system,mainly includes lane keeping,lane departure warning,vehicle lateral assistant power control,etc.Its application can greatly reduce traffic accidents caused by driver misoperation or fatigue.Among them,considering the lateral power assistant control of the vehicle,the existing research mostly ignores the difference of the driver's operational behavior,and designs the assitant control system of the vehicle with fixed parameters,resulting in the driver's acceptance of the assistant system is reduced.Therefore,for the driver's steering characteristics,designing a lateral assistant power control method that more satisfies the driver's needs has practical significance for improving the adaptability of the vehicle lateral assistant system and vehicle driving stability.Based on the Cyber Physics System(CPS)theory,this paper studies the lateral assistant power control method for the typical cyber physical fusion system of vehicle lateral control composed by human-vehicle-road system.Firstly,based on the existing lateral driver model,the Gradient Correction parameter estimation method of the driver model is designed,and the experiment platform of driving simulation is used to estimate the parameters of the driver model.On this basis,this paper designs the vehicle lateral assistant power T-S fuzzy control method considering the driver's operating characteristics.At the same time,according to the driver's low acceptance of the existing assistant system,a method of Model Predictive Control(MPC)considering driver's satisfaction is proposed.Research work of this paper includes the following three aspects:Firstly,due to the difficulty of the driver's physiological parameter detection and the large interference of the detection equipment to the driver,this paper designs the driver model parameter estimation method based on the Gradient Correction method.The advantages of the Gradient Correction estimation method are analyzed and compared.The driving simulation platform is used to verify the driver model obtained by the estimation method.The proposed model can reflect the driver's operation and verify the effectiveness of the method.Secondly,for the problem of fixed model parameters in the current lateral assistant power control,after the driver model parameters are estimated by Gradient Correction,and the assistant controller is designed.Considering the driver model in the human-vehicle-road closed-loop control system,a vehicle lateral assistant control system that considers the driver's steering characteristics is designed.At the same time,considering the different assistant needs of the driver due to different reaction time,an assistant control method considering the total reaction time of the driver is designed.The experiment proves that the proposed control method can ensure driving safety.Finally,for the problem of assistant control system and the driver to jointly control the vehicle,driving safety and driver satisfaction.Firstly,the definition of driver's satisfaction is determined,and the satisfaction function of the comprehensive assistant torque and the change rate of assistant torque is established.The optimization target considering driver's satisfaction and the vehicle road tracking accuracy is established.At the same time,Model Predictive Control method is used to design the assistant control system with the limit of the vehicle of yaw angle and lateral movement,and the magnitude and the change rate of assistant torque,and the driving safety as the constraint.Finally,the experiment proves that the proposed method can ensure the driver's satisfaction with the assistant control system under the condition of safe driving,and improve the coordination of human-vehicle while ensuring the accuracy of road tracking.