基于遗传算法的自动泊车控制方法研究  

Research on automatic parking control method based on genetic algorithm

作　　者：王靖岳[1,2] 王若羽 王浩天 WANG Jingyue;WANG Ruoyu;WANG Haotian(School of Automobile and Transportation,Shenyang Ligong University,Shenyang 110159,Liaoning,China;State Key Laboratory of Automotive Simulation and Control,Jilin University,Changchun 130025,Jilin,China;School of Automation,Shenyang Aerospace University,Shenyang 110136,Liaoning,China)

机构地区：[1]沈阳理工大学汽车与交通学院,辽宁省沈阳市110159 [2]吉林大学汽车仿真与控制国家重点实验室,吉林省长春市130025 [3]沈阳航空航天大学自动化学院,辽宁省沈阳市110136

出　　处：《农业装备与车辆工程》2023年第3期11-15,共5页Agricultural Equipment & Vehicle Engineering

基　　金：汽车仿真与控制国家重点实验室开放基金资助(20191203);国家重点研发计划项目(2018YFB0105004)。

摘　　要：针对停车难问题提出一种基于遗传算法的自动泊车控制方法。建立了车辆低速运动学模型,并针对平行泊车这一类型,规划两圆弧相切泊车路径,选取9次多项式拟合规划的路径,解决曲率突变问题。搭建了平行泊车模糊控制器,并利用遗传算法优化其隶属度函数。仿真结果表明,车辆能跟踪之前规划的泊车路径,并且在不同工况下均能完成自动泊车操作。Aiming at the current problem of difficult parking,an automatic parking control method based on genetic algorithm was proposed.Firstly,a low-speed kinematics model of the vehicle was established,and for the type of parallel parking,two arcs tangent parking paths were planned.The ninth degree polynomial fit planned path was selected to solve the problem of sudden curvature change,then a parallel parking fuzzy controller was built,and the genetic algorithm was used to optimize its membership function.Simulation results show that the vehicle can track the previously planned parking path,and in other different tasks,automatic parking operation can be completed under all conditions.

关 键 词：自动泊车 路径规划 模糊控制 遗传算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] U463.6[自动化与计算机技术—计算机科学与技术]