概率路线图环境下自动驾驶路径规划优化
Optimization of Automatic Driving Path Planning in Probabilistic Roadmap Environment
周俊霄 1唐晓峰1
作者信息
- 1. 扬州大学机械工程学院,江苏 扬州 225127
- 折叠
摘要
针对传统概率路线图算法不能满足自动驾驶车辆在雨雪天气路面湿滑的情况下安全行驶的特定场景需求,提出了基于概率路线图算法对其构型采样和碰撞检测进行优化的新型概率路线图算法.基于环境建立栅格地图,优化传统概率路线图算法,在构型采样和碰撞检测阶段加入距离检测来控制采样点和连接边线远离障碍物,改变最终的无向图,避免车辆紧贴障碍物路径前进的危险情况,提高了车辆行驶的安全性.结果表明,该算法可以准确选择路径,达到了预期效果.
Abstract
Aiming at the specific scenario that the traditional probabilistic roadmap algorithm cannot meet the needs of safe driving of self-driving vehicles on slippery roads in rainy and snowy weather,a new probabilistic roadmap algorithm based on the probabilistic roadmap algorithm is proposed to optimise its configuration sampling and collision detection.A raster map is established based on the environment,and the traditional probabilistic roadmap algorithm is optimised,and distance detection is added to the configuration sampling and collision detection phases to control the sampling points and connecting edges away from the obstacles,so as to change the final undirected graph,avoid the dangerous situation of the vehicle advancing along the path close to the obstacles,and improve the safety of the vehicle travelling.Simulation results show that the algorithm can accurately select the path and achieve the expected results.
关键词
路径规划/轨迹优化/概率路线图法/自动驾驶Key words
path planning/trajectory optimization/probabilistic roadmap method/autonomous driving引用本文复制引用
基金项目
扬州大学省级大学生创新创业训练计划项目基金(131101228)
创业训练项目(202311117166T)
出版年
2024
NETL
NSTL
万方数据