Font Size: a A A

Research On Path Planning And Tracking Of Vehicle Vertical Parking System

Posted on:2020-06-07Degree:MasterType:Thesis
Country:ChinaCandidate:G YinFull Text:PDF
GTID:2392330590487381Subject:Vehicle engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the rapid development of the economy and the improvement of people's living standards,cars have gradually entered the homes of ordinary people.As the number of car ownership increases year by year whether it is the major commercial plazas or the parking spaces in the residential quarters,the parking spaces are becoming more and more crowded,and this kind of parking difficulty often causes urban congestion and serious traffic accidents.Through the study of automatic parking at home and abroad,this paper chooses path planning and path tracking as the research route of text.In this dissertation,the vehicle is first studied under low speed conditions.By means of the Ackerman steering principle,the unequal left and right front wheel angles are simplified to an equivalent front wheel angle,and the feasibility of the vehicle kinematics model established in Simulink is verified by comparing the simulation results of the Carsim model with the same vehicle parameters.Secondly,the path planning is carried out for single-step parking under ideal conditions and multi-step parking in narrow parking environment.Considering the obstacle avoidance conditions during parking,the starting area of parking is established.Then,by selecting different starting points in the parking starting area,the relevant parameters are added into the Simulink kinematics model,and the motion trajectory points of the center point and surrounding vertices of the rear axle of the vehicle are obtained by running the simulation.The safety of automatic parking and the length of the path are the key to evaluate the parking function.This paper selects the particle swarm algorithm that does not have too high requirement for the target optimization function to optimize the safety of the parking and the total distance of the trajectory.By analyzing the corresponding constraints and establishing the constraint equation,the penalty function method is used to transform the linear and nonlinear constraints into unconstrained optimization problems,and the particle swarm optimization is used to obtain a reasonable multi-step parking trajectory.Then the Matlab and Simulink joint simulation model is established.By adjusting the predicted time domain Np value and controlling the time domain Nc value,the error of the trajectory tracking is kept within a reasonable range.Then,in the MPC model predictive control algorithm,the vehicle equivalent front wheel angle constraint and speed constraint and their respective incremental constraints are established,and the typical three kinds of trajectories in the vertical parking process which are respectively: the backward circular trajectory and the forward circular trajectory and the backward linear trajectory are tracked.Finally,the Carsim software is used to obtain the motion animation of the vehicle under three kinds of trajectories,which verifies the feasibility of the parking trajectory tracking controller.
Keywords/Search Tags:Automatic parking, Path planning and tracking, Model predictive control, Carsim/Simulink co-simulation
PDF Full Text Request
Related items