论文部分内容阅读
针对人与机器人共存的服务空间中人的定位跟踪问题,提出了一种基于分布式激光雷达协作感知的全局定位跟踪方法。分别对各台激光雷达获取的数据进行统计检验,将其分为静态数据和动态数据,利用静态数据完成各台激光雷达的位置标定,实现背景消除。动态数据通过无线网络传送到服务器,将来自同一时刻不同激光雷达的动态数据组成观测数据的一帧,实现数据的同步与融合。对获取的每帧数据进行基于迭代最近点算法的轮廓模型匹配,区分各个目标。采用基于位置-速度的关联门对相邻两帧的检测目标进行关联,实现对各动态目标的跟踪。实验验证了该方法在解决人的定位与跟踪问题的有效性,与基于视觉的定位跟踪方法相比,本系统在定位精度和跟踪成功率上优势明显。
In order to solve the problem of locating and tracking people in the service space coexisting with robots, a global positioning and tracking method based on cooperative lidar of distributed lidar is proposed. The data obtained from each laser radar are statistically tested, which are divided into static data and dynamic data. The static data are used to complete the position calibration of each laser radar to eliminate the background. The dynamic data is transmitted to the server through the wireless network, and the dynamic data from different lidar at the same time form a frame of observation data to synchronize and fuse the data. For each frame of data obtained, iterative nearest-neighbor algorithm based on the contour model matching, to distinguish between the various goals. Based on the position-velocity association door, the detection targets of two adjacent frames are correlated to track each dynamic target. Experiments show that this method is effective in solving the problem of locating and tracking people. Compared with the method of locating and tracking based on vision, this method has obvious advantages in locating accuracy and tracking success rate.