A Study On Path Planning And Tracking Control Strategy For Automatic Parking System

The automatic parking system is one of the representative systems of the intelligent time of the vehicle,and has become the main way for major auto companies to demonstrate their strength in vehicle intelligence.In today’s increasingly scarce parking spaces and the ever-decreasing parking space,the automatic parking system has gradually become a “standard accessory” for vehicles and has become one of the major reference items for consumers to purchase cars.This article focuses on the system control algorithm,which is the core issue of the automatic parking system.Divided into two aspects of path planning and path tracking for discussion and analysis.In terms of path planning,double constant speed trajectory was proposed based on the low-speed motion characteristics of the vehicle;Based on the manipulation behavior of the driver in parking conditions,the DCD trajectory,which is mainly applied to the parking path planning,was further proposed;supplement and description of DCD trajectory related parameters,summarized the use of DCD trajectory to plan parallel parking and vertical parking paths,used the method of reverse path planning to analyze the problem of the minimum feasible parking space and parking feasibility in the case of one-time parking.In terms of path tracking,this paper studied the influence of the double speed variation on the double constant speed trajectory,and finds that the double constant speed trajectory has a fixed shape when the two speeds of the vehicle are simultaneously and proportionally changed;the characteristic of the double constant speed trajectory was further extended to the DCD trajectory,and the special property of the DCD trajectory was obtained;based on the special property of DCD trajectory,a steering wheel angle control algorithm based on real-time vehicle speed and vehicle stage is designed,this algorithm uses the characteristics of DCD trajectory to find out the relationship between vehicle speed and steering wheel speed,which significantly reduced the impact of vehicle speed fluctuation on path tracking accuracy;since the parking control algorithm based on the DCD trajectory characteristics is an open-loop algorithm,the impact of random interference on the path tracking effect cannot be dealt with,in order to improve the stability of the algorithm,this paper derived the steering wheel angle control rate of the vehicle following a straight line target trajectory based on the non-time based path tracking control theory,added a closed-loop algorithm to adjust the position of the vehicle in the original open-loop algorithm,used Carsim’s vehicle model to verify the theoretical feasibility of the improved "half closed loop" algorithm.The biggest characteristic of this paper found the vehicle DCD trajectory,which is the combination of parking path planning and path tracking.For path planning,the system does not need to know the specific value of each point of the trajectory.It only needs to know the yaw angle of key points on the parking path to complete the parking path planning,which greatly reduces the amount of calculation of the system.For parking path tracking,the control algorithm is mostly logical judgment,and a simple algorithm can greatly improve the system’s working efficiency.The above algorithm is still in the theoretical stage.In the later stage,relevant research certification on real vehicles is required.