摘要
为了完成越野环境中无人自主导航车的障碍检测任务,应用四线激光雷达,提出了1种新的障碍检测算法,该算法利用检测区域的坡度信息进行障碍检测,包括坐标变换、候选障碍点提取、候选障碍点聚类及点簇高度求取4个步骤。为了克服激光雷达检测盲区与抑止测量过程中干扰噪声的影响,运用了卡尔曼滤波算法对目标障碍进行了滤波处理。试验结果表明,障碍检测算法稳定可靠。
In order to obtain obstacle information in a cross-country environment for an autonomous land vehicle (ALV), a novel obstacle detection algorithm is proposed by using a four-layer laser radar LD_ML. The obstacle detection algorithm achieves the obstacle detection task by the gradient information of detection regions in four steps: ( 1 ) changing scanning points' coordinate system from the LD_ML to the ALV; (2) extracting candidate obstacles points from LD_ML range data; (3) clustering the candidate obstacle points; (4) obtaining the height information of candidate obstacle clustering. To overcome the blind detecting areas and the measurement noise of the LD_ML, Kalman filtering technology is used to estimate the position of obstacles and to track the obstacles. The experimental results show that the obstacle detection algorithm is reliable and robust.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2006年第5期618-621,625,共5页
Journal of Nanjing University of Science and Technology
关键词
激光雷达
越野
障碍检测
障碍跟踪
laser radar
cross-country
obstacle detection
obstacle tracking