Research On Automatic Parking Path Planning And Motion Control

As convenient and speedy means of transportation,autos have become a significant part of people’s life.The number of vehicles has grown on a yearly basis,which causes sorts of problems like the insufficiency of parking spaces,small dimensions of parking spaces and difficult parking.The autonomous parking system has become a research focus,which is capable of solving these problems efficiently and relieving drivers parking burden by autonomous parking spaces researching and autonomous parking.This paper studies autonomous parking,path tracking,motion control of unmanned vehicle relied on project “Multiscale networked vehicle collaborative service platform(2018YFB1600701)”.The principal content of this paper including:First of all,studying the strategy method of unmanned vehicle’s autonomous parking,conducting the abstract analysis about common parking spaces,establishing and analyzing the parking space simulation models based on software Pre Scan? analyze a series of possible driving operations in the case of automatic parking and summarize the driving of automatic driving vehicles for AVP service.The basic driving behavior models on the scene,combined with the finite state machine to establish unmanned vehicle’s autonomous parking decision models,and using the Stateflow module in Simulink to build a state machine model to gradually make autonomous parking vehicle decision on automatic parking scenarios.Secondly,studying the path planning method of unmanned vehicle’s autonomous parking.This paper considers that the single A~* algorithm and the artificial potential field method have insufficient problems when planning the vehicle path,proposes a path planning method that combines the two,and adopts the method of path discretization to design global path plan based on the combination of artificial potential field algorithm and A~* algorithm.The method combines the advantages of the A~* algorithm and the artificial potential field method,and sets the conditions to switch the algorithm to avoid the shortcomings of the single algorithm.In view of the problem that the trajectory planned by the fusion algorithm does not fully meet the vehicle operating conditions,considering the kinematic constraints of the vehicle model,a global path optimization method is designed based on the Dubins curve? A vertical parking trajectory planning method based on Reeds-Shepp curve design.The effectiveness of automatic parking trajectory planning is verified by simulation.At last,studying the motion control method of unmanned vehicle’s autonomous parking based on model predictive control algorithm.This paper analyzes vehicle dynamics system and establishes a 3 DOF vehicle dynamics model,designed the control algorithm,including the design performance function preview model,proposes control constraints,so that the automatic parking motion controller is designed.Using MPC controller and the PID controller to simulate and verify the double-shift line trajectory,the parking-seeking trajectory and the vertical parking trajectory respectively.The performance of the two controllers was compared to verify the effectiveness and trajectory of the motion control method studied in this paper.