摘要
动态环境下运动物体跟踪是移动机器人研究的难点之一;文章提出了一种基于激光雷达的自主动态障碍检测与跟踪方法;该方法首先利用最近邻聚类法将环境数据聚类为不同的障碍物;然后利用最近邻特征匹配算法关联相邻两帧的障碍物;最后提出一种新的基于障碍物时空关联性分析的的障碍物动静态识别算法,并采用α-β滤波算法对动态障碍的位置和速度进行了估计;利用机器人平台对该方法进行验证,实验结果表明了其有效性。
Detection and tracking of moving object is a difficult problem in mobile robot research. An autonomous approach for detection and tracking of moving obstacles is presented using 2D laser scanner. It based on the character--matching of obstacles clustered from environment data. The types of obstacles are determined with the analysis of spatiotemporal association, and implements the estimate of moving object using the α-β algorithm finally. The experiment shows that the method can accomplish the task effectively.
出处
《计算机测量与控制》
CSCD
北大核心
2012年第3期816-819,共4页
Computer Measurement &Control
关键词
聚类
障碍物关联
时空关联性分析
动态障碍物跟踪
clustering
obstacle registration
the analysis of spatiotemporal association
tracking of moving object
