集中交互式多传感器联合概率数据互联算法

为了解决杂波环境下多传感器多机动目标跟踪问题,本文提出了一种集中交互式多传感器联合概率数据互联算法。本文提出的算法首先应用广义S-D分配的规则对每个传感器送来的观测数据进行排列组合,并对所有的测量组合进行有效性判断,然后应用数据压缩的方法将每个有效量测组合压缩成一个等效量测点并根据每个等效量测点的联合似然函数计算其联合互联概率,最后在此基础上应用交互式多模型算法的思想以处理目标出现机动的问题。本文最后给出了该算法的分析,仿真结果表明,本文算法能够很好地解决杂波环境下多传感器多机动目标的跟踪问题。

Centralized interacted multi-sensor joint probabilistic data association algorithm

In order to resolve the problem of tracking multiple maneuvering targets in centralized multi-sensor situation, a new centralized interacted multi-sensor joint probabilistic data association (CIMM-MSJPDA) algorithm is presented, which is a combination of centralized Multi-sensor Joint Probabilistic Data Association (MSJPDA) techniques and Interacted Multiplied Model (IMM) method. The algorithm is proposed to associate measurements from each sensor with the target from which the measurements originated by use of MSJPDA techniques, and to track maneuvering targets with IMM method. In the new algorithm, the measurements from each sensor are permuted and combined, and some combinations will be accepted according to a rule. Then, all of the measurements in each combination accepted will be compacted into an equivalent measurement which is associated with targets by using the MSJPDA techniques. Finally, the interacted multiple model method is applied to resolve the maneuvering target tracking problem. Simulations are designed to compare the performance of the new algorithm with a centralized MSJPDA method and IMMJPDA algorithm. The results show that the use of CIMM-MSJPDA algorithm enables the centralized multi-sensor system track multiplied maneuvering targets and its tracking performance is much better than that of IMMJPDA algorithm.