分布式多目标自动跟踪和特征管理算法

一种可扩展的分布式多目标跟踪和特征管理的算法(DMTIM),适用于对多目标进行跟踪的传感器网络。DMTIM由数据关联、多目标跟踪、特征管理,以及信息融合所组成。数据关联和未知数量多目标的跟踪通过马尔科夫蒙特卡罗数据关联实现。通过信息融合来获取所有相邻传感器的本地一致性,实现特征管理。试验证明,DMTIM算法能够通过传感器网络实现对多目标有效的跟踪和特征管理。

Distributed Multiple-Target Tracking and Identity Management Algorithm

In this paper, we pay much afford on the researching of a scalable distributed multiple-target tracking and identity management (DMTIM) algorithm that can track multiple targets and manage their identities efficiently in a distributed sensor network environment. DMTIM finds a globally consistent solution by maintaining local consistency among neighboring sensors. The data association and multiple-target tracking problems are efficiently solved by Markov chain Monte Carlo data association (MCMCDA). DMTIM manages identities of targets by incorporating local information and maintains local consistency among neighboring sensors through information fusion.