Research On Lane Keeping System Based On Improved Artificial Potential Field Method

With the rapid development of the automobile industry,the pace of people's life become faster,many drivers are not good at driving experience,some drivers do not have good driving habits.The traffic accidents are common because of fatigue driving and driving without mind.The development of advanced driver assistance systems has become increasingly important,It can improve people's driving pleasure,avoid more traffic accidents,and protect the safety of occupants.ADAS has many driver assistance functions,LKS(Lane Keeping System)is one of the system functions.In this paper,we deeply research on the lane keeping control algorithm at the aspects of theory and experiment.Firstly,the paper introduces that the trend of intelligent automobile has become unstoppable,and research at the aspects of theory and experiment has important significance and the practical value.The areas and technologies related to LKS are introduced.At the same time,the status of research about lane keeping control algorithm and its HIL(Hardware in the Loop)test platform is summarized.In the MATLAB / Simulink,the driver model and the vehicle-to-road relative position model are established.The vehicle parameters,the simulation conditions and the connection with Simulink are set up in CarSim.We establish the closed loop system and simulate and analyze it.It provides a simulation and test platform for improving and verifying control algorithms.Secondly,On the basis of with no more vehicle mechanism,in order to realize the automatic steering function of the lane keeping system,the motor in EPS is used as the steering actuator.In order to reasonably arrange the work of the motor,and prevent the the conflicts between two modes,we design the coordinated control for the motor.then we introduce working principle and control strategy of the motor under active steering mode.Thirdly,the problem of tracking accuracy of the lane keeping system with deviation feedback PID control is not high,so an artificial potential field method is proposed.Through the comparison of the two methods from the simulation,the artificial potential could improve the accuracy of the vehicle route tracking.Then we analyze that there are some influence on the construction of the potential field in the aspect of vehicle speed.In order to avoid the large deviation of the vehicle,the lanesafety boundary line is proposed,and the repulsive field of the safety boundary line is added to the potential field,which further improves the safety and the effect of the LKS for vehicle.Thus,a new potential field is established that includes the speed of the vehicle and the boundary line.Finally,in order to verify the effectiveness of the proposed control algorithm and compare the control effects of several control algorithms,we carry out Simulink and CarSim joint simulation,and build a hardware-in-the-loop experiment platform.The simulation and experiment's results show that the correctness of coordinated control of working mode of motor,and the improved potential field method improves the tracking accuracy of the lane keeping system.It satisfies the real-time performance of the lane keeping system and improves the tracking effect.