自动驾驶路径跟踪中实时动态障碍物处理研究  

Research on real time dynamic obstacle handling in autonomous driving path tracking

作　　者：闫茜 梁方英[1] Yan Xi;Liang Fangying(Zhejiang Institute of Mechanical&Electrical Engineering,Hangzhou 310051,China)

机构地区：[1]浙江机电职业技术学院,浙江杭州310051

出　　处：《汽车知识》2024年第1期107-109,共3页Auto Know

摘　　要：随着自动驾驶技术的不断发展,路径跟踪中实时动态障碍物的处理成为关键挑战。本文回顾了自动驾驶路径规划技术的演进,重点研究了优化传感器数据融合、强化目标检测与跟踪、提高障碍物分类准确性、创新实时地图更新策略,以及强化路径规划与实时动态障碍物协同等五大改进策略。这些策略的综合应用提高了自动驾驶系统对复杂交通环境的适应性和安全性,为实现可靠的路径规划奠定了基础。With the development of autopilot technology,the real-time dynamic obstacle processing in path tracking becomes a key challenge.This paper reviews the evolution of autopilot path planning technology,this paper focuses on five improvement strategies,such as optimizing sensor data fusion,strengthening target detection and tracking,improving the accuracy of obstacle classification,innovating real-time map updating strategy,and strengthening the coordination between path planning and real-time dynamic obstacle.The comprehensive application of these strategies improves the adaptability and security of the automatic driving system to the complex traffic environment,and lays a foundation for reliable path planning.

关 键 词：车辆自动驾驶路径跟踪 实时动态障碍物处理 改进措施 

分 类 号：U463.6[机械工程—车辆工程]