Poisson Multi-Bernoulli Mixture Filter for Heavy-tailed Process and Measurement Noises

A novel Poisson multi-Bernoulli mixture (PMBM) filter is proposed to track multiple targets in the presence of heavy-tailed process and measurement noises. Unlike the standard PMBM filter that requires the Gaussian process and measurement noises, the proposed filter uses the Student’s t distribution to model the heavy-tailed noise feature. It propagates Student’s t-based PMBM posterior in the closed-form recursion. The introduction of the moment matching method enables the proposed filter to deal with the process and measurement noises with different heavy-tailed degrees to some extent. Simulation results demonstrate that the overall performance of the proposed filter is better than the existing heavy-tailed noise filters in various scenarios.