Abstract: Probabilistic multi-hypothesis tracker (PMHT) algorithm is an important data association algorithm owing to the fact that its computational cost is linear with respect to the number of targets and observations. However, the present PMHT is a measurement oriented algorithm which is hospitable to clutters. Therefore, this article proposes a target oriented probabilistic multi-hypothesis tracker (TO/PMHT), which takes into account the distances between measures and targets, and calculate the true association probabilities. Thus, the weakness of PMHT of hospitality to clutters is greatly reduced. A comparison of the proposed TO/PMHT with measurement oriented PMHT and traditional data association algorithms, such as multi-hypothesis tracker (MHT) by computer simulation shows that TO/PMHT increases tracking performance and computation efficiency greatly.

Key words: probabilistic multi-hypothesis tracker, data association, target oriented, tracking, multi-hypothesis tracker