交通运输系统工程与信息

交通运输系统工程与信息 ›› 2022, Vol. 22 ›› Issue (4): 293-301.DOI: 10.16097/j.cnki.1009-6744.2022.04.033

• 智能车联网技术与应用 • 上一篇    下一篇

面向动静态混合环境的智能车运动规划方法

齐尧，朱彦齐，李永乐*，徐友春   

  1. 陆军军事交通学院，天津 300161
  • 收稿日期:2021-12-28 修回日期:2022-03-02 接受日期:2022-03-17 出版日期:2022-08-25 发布日期:2022-08-22
  • 作者简介:齐尧(1994- )，男，四川仪陇人，博士生。
  • 基金资助:
    军队学科专业建设项目

Motion Planning Method for Intelligent Vehicles Under Constrained Dynamic and Static Environment

QI Yao, ZHU Yan-qi, LI Yong-le* , XU You-chun   

  1. The Army Military Transportation University, Tianjin 300161, China
  • Received:2021-12-28 Revised:2022-03-02 Accepted:2022-03-17 Online:2022-08-25 Published:2022-08-22
  • Supported by:
    Program of Army Subject Specialty Construction(4142Z17)。

PDF

摘要： 针对存在动态和静态障碍物环境中的智能车运动规划问题，本文提出一种基于三维搜索的方法。该方法首先在笛卡尔坐标系中增加时间维度以建立时间-空间栅格，构建不同车速下的车辆运动基元；将智能车转化为多个圆心组成的线段，采用膨胀静态障碍物的方法进行不规则障碍物碰撞检测，运用时间间隔法将动态障碍物碰撞检测简化为线段交叉性检测；构建包含最大速度和加速度约束的速度启发式函数，用于引导搜索树在时空空间中快速到达目标位置和速度；最后基于启发式方法和局部运动规划方法在时空栅格中进行搜索，获取融合时间信息和“停止-等待”等决策信息的运动轨迹。实验结果表明：本文运动规划方法在动静态混合环境中能够引导智能车安全行驶，相比速度障碍方法，安全行驶的平均成功率提升23%；相比混合A*方法，平均成功率提升19%，行驶耗时缩短21%。

关键词: 智能交通, 智能车, 搜索算法, 运动规划, 动态环境

Abstract: Focusing on intelligent vehicle motion planning in an environment with dynamic and static obstacles, a search-based method in 3D space is proposed. The method establishes the spatiotemporal grid by adding a time dimension to the Cartesian coordinate system and constructs vehicle motion primitives at different speeds. And an intelligent vehicle is transformed into a line segment composed of multiple circle centers. The method then uses the expanding static obstacle method for rapid collision detection of irregular obstacles and simplifies dynamic obstacle collision detection to line segment intersection detection by the time interval method. For guiding the search tree to reach the target position and velocity quickly in spatiotemporal space, a velocity heuristic function constrained by maximum velocity and acceleration is constructed. Finally, the motion trajectory of integrating time information and “stop-wait”decision information is obtained by searching in the spatiotemporal grid, which is based on a heuristic method and partial motion planning. Experimental results show that the proposed motion planning method can guide the intelligent vehicle to drive safely in a constrained dynamic and static environment. The average success rate of safe driving is increased by 23% compared with velocity obstacles. Compared with hybrid state A*, the average success rate is increased by 19%, and the total driving time is reduced by 21%.

Key words: intelligent transportation, intelligent vehicle, search-based algorithm, motion planning, dynamic environment

中图分类号: