基于航迹隶属度的分布式系统数据融合算法

Distributed System Data Fusion Algorithm Based on Track Fuzzy Membership

航迹关联与航迹融合是分布式目标跟踪系统数据融合的关键。本文研究了基于航迹隶属度的数据融合算法。综合各传感器航迹估计形成的目标运动状态特征向量与传感器分辨率,根据模糊聚类算法建立各观测时刻航迹隶属度矩阵与系统航迹关联决策矩阵,解决融合中心航迹关联问题。根据加权融合算法思想,结合各观测时刻航迹隶属度矩阵,实时、动态分配航迹号集合中各局部航迹权值,解决目标航迹融合问题。蒙特卡罗仿真表明,算法航迹关联效果明显优于加权航迹关联算法,并得到与简单航迹融合算法一致的目标融合航迹。

Track association and track fusion are the key problems of data fusion in distributed target tracking system.A data fusion algorithm based on track fuzzy membership is studied.By integrating the moving estimation character vector provided by track estimation of each local sensor and the performance of sensor resolution,track fuzzy membership matrix and systemic track association decision matrix at every measurement time are established according to Fuzzy Clustering Method（FCM）,then problem of track association in fusion center is solved.Combining with the idea of weighted fusion algorithm and track fuzzy membership matrix at every measurement time,real time and dynamic weighting factors of all local tracks in track aggregation are alloted,and problem of track fusion is solved.Monte Carlo simulation shows that proposed algorithm is superior to weighted track association algorithm,and tracking system can obtain the target fusion track that have same tracking performance of Simple Fusion（SF） algorithm.