A Particle Multi-Target Tracker for Superpositional Measurements using Labeled Random Finite Sets

摘要

In this paper we present a general solution for multi-target tracking withsuperpositional measurements. Measurements that are functions of the sum of thecontributions of the targets present in the surveillance area are calledsuperpositional measurements. We base our modelling on Labeled Random FiniteSet (RFS) in order to jointly estimate the number of targets and theirtrajectories. This modelling leads to a labeled version of Mahler'smulti-target Bayes filter. However, a straightforward implementation of thistracker using Sequential Monte Carlo (SMC) methods is not feasible due to thedifficulties of sampling in high dimensional spaces. We propose an efficientmulti-target sampling strategy based on Superpositional Approximate CPHD(SA-CPHD) filter and the recently introduced Labeled Multi-Bernoulli (LMB) andVo-Vo densities. The applicability of the proposed approach is verified throughsimulation in a challenging radar application with closely spaced targets andlow signal-to-noise ratio.