采用改进人工势场法的动态无人车路径规划  

Unmanned Vehicle Path Planning Based on Structured Road Improved Artificial Potential Field Method

作　　者：邱朋 汪光 赵理[1] 赵博阳 QIU Peng;WANG Guang;ZHAO Li;ZHAO Bo-yang(School of Mechanical and Electrical Engineering,Beijing Information Science and Technology University,Beijing 100192,China;School of Artificial Intelligence,Hebei Univeristy of Technology,Tianjin 300401,China)

机构地区：[1]北京信息科技大学机电工程学院,北京100192 [2]河北工业大学人工智能与数据科学学院,天津300401

出　　处：《机械设计与制造》2023年第3期291-296,共6页Machinery Design & Manufacture

基　　金：国家自然科学基金面上项目(51575044);北京市教委面上项目(KM201811232003)。

摘　　要：针对无人车在动态转向避障过程中的,基于结构化道路改进人工势场模型进行了转向避障路径规划。首先应用椭圆化距离代替斥力势场的中的实际距离,其次,引入道路边界斥力势场模型,从而在较小车道空间内获得局部避障路径,为了更加准确的描述无人车行驶环境及动态障碍物信息,引入障碍物速度斥力势场。最后,建立了以转角为控制变量的纯追踪算法模型进行路径跟踪。Matlab仿真结果表明,对比传统人工势场法,应用改进的人工势场模型能够获得平顺且安全的局部避障路径,纯追踪控制器模型具有良好的路径跟踪性能。In order to ensure the active safety of unmanned vehicles in the process of steering obstacle avoidance,a path planning for steering obstacle avoidance was carried out based on a structured road improvement artificial potential field model. First,the elliptic distance is used to replace the actual distance in the repulsive potential field. Second,the road boundary repulsive potential field model is introduced to obtain a local obstacle avoidance path in a small lane space. In order to more accurately describe the driving environment and dynamics of unmanned vehicles Obstacle information,introducing the obstacle’s velocity repulsive potential field. Finally,a pure tracking algorithm model with the previous turning angle as the control variable is established for path tracking. Matlab simulation results show that the improved artificial potential field model can obtain smooth and safe local obstacle avoidance paths,and the pure tracking controller model has good path tracking performance.

关 键 词：人工势场法 路径规划 纯追踪算法 路径跟踪 

分 类 号：TH16[机械工程—机械制造及自动化] U461.91[机械工程—车辆工程]