摘要
当传感器的测量精度不同时,采用最近邻算法进行量测分配往往会得出错误的关联结果;包含有不同数量的点迹的航迹间的关联,由于样本数量不同,不能用样本方差来判断航迹间的相似性。利用数理统计方法实现对测量精度不同,且包含的点迹数也不相同的航迹间的关联判别。假定传感器的测量误差是统计独立的,服从零均值和恒定方差的高斯分布,利用最大似然概率求解不等样本容量的量测间的相似性度量,可以作为包含有不同数量的点迹的航迹间的关联指标。Mat-Lab仿真表明,只要测量样本量大于5就可以实现正确关联,算法的实时性可以接受;正确关联率主要与目标的密集度有关。
There is no clear means for defining what constitutes ＂nearest neighbors＂ among measurements and tracks with different error covariance.When the number of measurements for each track is unequal,the track with the minimum squared error is not sure to be the correct association.A decision rule is developed for associating a set of tracks with unequal measurements using mathematical statistics approach,under the assumption that the measurement errors are independent and Gaussian distributed with zero mean and constant variance.The algorithm is tested with Matlab simulated data.Correct association can be made when the number of measurements of each track is more than 5.The main influencing factor of association is the closeness of tracks one another.
出处
《中国电子科学研究院学报》
2013年第3期255-257,265,共4页
Journal of China Academy of Electronics and Information Technology
基金
总装预研项目(102060302)
关键词
多目标跟踪
航迹关联
数理统计
雷达
Multiple-Target Tracking
Track Correlation
Mathematical Statistics
Radar
