基于ET-GMPHD算法的编队目标跟踪方法

针对低检测环境下编队目标的跟踪问题,提出了一种基于扩展目标高斯混合概率假设密度(extended target Gaussian mix probability hypothesis density, ET-GMPHD)算法的编队目标跟踪方法。首先,保留修剪掉高斯项的同时将其外推,用Jensen-Shannon(JS)散度衡量下一时刻状态估计值与外推值间的相似程度,并以此反映是否有目标丢失,保证真实目标不被修剪,解决了因目标漏检导致跟踪结果不准确的问题。其次,结合循环阈值聚合法得到编队整体的状态估计,消除了估计状态集合中状态值过多造成的影响。最后,仿真实验表明,该方法能够在检测概率极低的情况下进行有效跟踪,并具有良好的跟踪性能。

Formation target tracking method based on ET-GMPHD algorithm

To solve the tracking problem of formation targets in low detection environment, a formation target tracking method based on the extended target Gaussian mix probability hypothesis density (ET-GMPHD) algorithm is proposed. Firstly, the Gaussian term is trimmed and extrapolated, and Jensen-Shannon (JS) divergence is used to measure the degree of similarity between the state estimate and the extrapolated value at the next moment, and to reflect whether there are target loss and guarantee the true targets are not trimmed, which solves the problem of inaccurate tracking results due to missed detection of the target. Secondly, combined with the cyclic threshold aggregation method, the state estimation of the overall formation is obtained, which eliminates the influence of excessive state values in the estimated state set. Finally, simulation experiments show that the method can effectively track with a very low detection probability and has a good tracking performance.