| 摘 要: 为了解决快速搜索随机树(RRT)算法存在收敛速度慢、生成路径质量差等问题,提出一种融合算法,即将人工势场法以及A* 算法的代价函数融合到RRT算法中。使用改进后的RRT算法生成路径,并将其所搜寻到的路径作为初始解传递给A* 算法,进而重新规划出一条更优路径,并结合车辆运动学约束,利用贝塞尔曲线平滑路径。选取两种地图场景进行仿真实验,实验结果表明,提出的改进策略可以更快地生成质量更高且更符合汽车实际行驶的路径。仿真结果证明了该算法能有效提升RRT算法的收敛速度和路径质量。 |
| 关键词: RRT算法;人工势场法;A* 算法;贝塞尔曲线 |
|
中图分类号: TP242.6
文献标识码: A
|
| 基金项目: 湖北省教育厅科学技术研究项目(B2022023) |
|
| Route Planning Strategy for Intelligent Vehicles Based on Optimized RRT Algorithm |
|
ZHOU Runmin, LIU Huaming, PENG Xitai
|
(Faculty of Automotive and Tra f f ic Engineering, Wuhan University of Science and Technology, Wuhan 430065, China)
2284373405@qq.com; liuhuaming@wust.edu.cn; pxitai624@gmail.com
|
| Abstract: Aiming at the slow convergence speed and poor quality of route generation of Rapidly-exploring Random Tree (RRT) algorithm, this paper proposes a fusion algorithm that integrates the artificial potential field method and the cost function of the A* algorithm into the RRT algorithm. The improved RRT algorithm is used to generate routes, and the routes searched are passed to the A* algorithm as the initial solution, which then replans a better path. Additionally, considering the kinematic constraints of the vehicle, Bezier curves are used to smooth the path. Two different map scenarios are selected for simulation experiments, and the results demonstrate that the proposed improved strategy can generate higher-quality routes more quickly, which are also more aligned with actual vehicle driving. The simulation results validate that the algorithm effectively enhances the convergence speed and route quality of the RRT algorithm. |
| Keywords: RRT algorithm; artificial potential field method; A* algorithm; Bezier curve |
