摘要
针对非结构道路负障碍物会对高速行驶的地面无人车辆造成车体侧倾、剧烈颠簸、轮胎深陷及爆胎等危险,提出了一种新的基于激光雷达的级联式非结构化道路负障碍物检测方法。首先,基于角度特征的广度优先搜索,可快速获得潜在的负障碍物,并同时完成负障碍物聚类工作;再结合环间距与高度特征对潜在负障碍物进行评价,最终提取负障碍物。实验结果表明:该方法在非结构化道路中等距离区间内具有良好的负障碍物检测性能。
Aiming at the problem that negative obstacles on unstructured roads can cause hazards such as vehicle body roll, violent bumps, deep tire sinking and tire blowouts to unmanned vehicles on the ground at high speeds, a new type of unstructured road based on LiDAR cascade negative obstacle detection method is proposed.Firstly, the breadth first search(BFS)based on angle features can quickly obtain potential negative obstacles, and at the same time complete the negative obstacle clustering work.Then, combine the ring spacing and height characteristics to evaluate potential negative obstacles.Finally, the negative obstacles are extracted.The experimental results show that this method has good negative obstacle detection performance in the middle distance interval of unstructured roads.
作者
朱建新
唐升平
辛涛
吴钪
ZHU Jianxin;TANG Shengping;XIN Tao;WU Kang(State Key Laboratory of High Performance Complex Manufacturing,Central South University,Changsha 410083,China;The National Enterprise R&D Center,Sunward Intelligent Equipment Co Ltd,Changsha 410100,China;Unit 32181 of the Chinese People’s Liberation Army,Xi’an 710032,China)
出处
《传感器与微系统》
CSCD
北大核心
2022年第11期130-133,142,共5页
Transducer and Microsystem Technologies
基金
国家重点研发计划资助项目(2019YFC1511503)
湖南省重点领域研发计划资助项目(2019SK2181)
湖湘青年英才项目(2019RS2053)。
关键词
负障碍物
非结构化道路
点云
negative obstacles
unstructured roads
point cloud
