Driver-Adaptive Lane Keeping Assistance Control

As an increasingly popular means of transportation,cars have brought great convenience to people's travel.In order to reduce the irreparable loss of life and property caused by traffic accidents,the problem of vehicle driving safety has become a key issue in the field of social transportation and automobile.As an important part of the active safety system,the Lane keeping assist(LKA)is essential for improving the lateral stability of the vehicle,reducing the driver's operating burden and the incidence of traffic accidents.Due to the differences in driver's operating habits and the different perceptions of lane departures,in order to meet the driving needs of different drivers,the LKAS system needs to be able to adaptively operate the driver to prevent the auxiliary system from colliding with the driver.Under the support of the National Key R&D Program(2016YFB0101102),“Electric Vehicle Intelligent Assisted Driving Technology R&D and Industrialization”,the specific research contents of this issue are as follows:Firstly,based on the simulation driving system to collect driving data,because the real data collection period is long and the cost is high,this paper uses PreScan to simulate the driver's actual demand scene.CarSim provides the vehicle dynamics model to build the simulation driving system.Firstly,the driving data of different drivers under different working conditions are collected by the simulated driver,the driver's lateral driving behavior characteristic parameters are analyzed,the driver's operation law in the lane keeping process and the key factors affecting the driver's operation are found.Secondly,the lane keeping auxiliary controller is designed based on the adaptive model predictive control theory.Due to the driver's operation,the uncertainty of the road environment and the nonlinearity of the driver model and the vehicle model,this paper designs the lane keeping auxiliary controller through adaptive model predictive control.By constructing a "human-vehicle-road" closed-loop model,the desired yaw rate is obtained;according to the vehicle's lateral deviation and deviation threshold algorithm,the intervention time of the auxiliary controller is accurately determined,and the deviation of the vehicle is prevented,so as to be expected.Finally,the reliability of the lane keeping assist system is verified based on the joint simulation of CarSim and Simulink.Through CarSim to build the vehicle model,road model and camera required for lane detection,the lane keeping auxiliary system is simulated and verified under linear and curve conditions,and the performance of the system is observed according to different offsets.