Set JPDA algorithm for tracking unordered sets of targets

In this article we show that traditional tracking algorithms should be adjusted when the objective is to recursively estimate an unordered (unlabeled) set of target state vectors, i.e., when it is not of importance to try to preserve target identities over time. We study scenarios where the number of targets is known, and propose a new version of the joint probabilistic data association (JPDA) filter that we call set JPDA (SJPDA). Simulations show that the new filter outperforms the JPDA in a two-target scenario when evaluated according to the mean optimal subpattern assignment (MOSPA) measure.

1 Introduction

Traditional tracking algorithms are tailored to the problem of tracking a—possibly unknown—number of targets with preserved target identity. That is, the interest is not only in estimating the states of each target, but also to keep track of the identities of the targets over time. Examples of such algorithms are the Probabilistic Data Association (PDA) and Joint PDA (JPDA) filters [1], [2], the Multiple Hypothesis Tracking (MHT) algorithm [3], [4], [5], [6], and particle filters [7], [8].