Research On Path Planning And Tracking Control Of Vehicle Parallel Parking System

In recent years,unmanned vehicles have developed rapidly.As the last step of unmanned vehicles,automatic parking systems have received extensive attention.The automatic parking system is of great significance for alleviating traffic difficulties,improving traffic efficiency,and reducing parking accidents.Based on the current research status of various aspects,this paper studies the parallel parking system from three aspects: path planning,path optimization and path tracking.The main contents are as follows:1.Based on the Ackerman steering foundation,a parking kinematics model at low speed was established.A simulation model was built in Simulink.Under the premise of the same vehicle parameters and input conditions,the trajectory output of the Simulink model and the Carsim model were compared and analyzed to verify the accuracy of the simplified kinematics model.2.Path planning: According to the parking space parameters and obstacle avoidance constraints during the parking process,the parking path was planned.For the ideal parking space,the two-arc tangent method is used to plan the single-step parking path.For the narrow parking space,the multi-step parallel parking path was planned by the reverse method.The planned path trajectory was simulated and verified in Matlab.3.Path optimization: The total length of the path plays a vital role in the efficiency of parking.This article takes the total length and safety of the path as the goal,and uses the particle swarm optimization algorithm to optimize the parking path.For the equality and inequality constraints in the parking process,the penalty function was used to transform the constrained optimization problem into an unconstrained optimization problem to solve,and the optimal path parameters that meet the optimization goals and constraints were obtained,A comparative analysis of the paths before and after optimization verifies the effectiveness of the particle swarm optimization algorithm.4.Path tracking: The model predictive control(MPC)was used to track and control the planned path,and the objective function is established with accuracy and stability as indexes.In order to prevent mutation phenomenon in the tracking process,the equivalent front wheel angle and speed and their respective incremental constraints were established,and the parking path tracking problem was transformed into a quadratic programming problem for solution.5.The joint simulation model of Carsim and Simulink was built,and the function of the designed model predictive controller was simulated and verified.The arc tangent single-step parking path and the multi-step parking path optimized by particle swarm optimization were tracked and controlled respectively.The error change and parking simulation animation in the tracking process are obtained,and the control effect of the controller was intuitively verified.