A Distributed Particle-PHD Filter with Arithmetic-Average PHD Fusion

TL;DR

This paper introduces a novel distributed particle-PHD filter that employs arithmetic-average fusion and efficient Gaussian mixture conversions, achieving high performance with low communication and computational costs in multi-target tracking.

Contribution

It presents a new distributed particle-PHD filtering approach using arithmetic-average fusion and significance-based GM pruning, improving efficiency and scalability.

Findings

Demonstrates excellent tracking performance in simulations.

Reduces communication and computation requirements.

Enables parallel processing of filtering and fusion steps.

Abstract

We propose a particle-based distributed PHD filter for tracking an unknown, time-varying number of targets. To reduce communication, the local PHD filters at neighboring sensors communicate Gaussian mixture (GM) parameters. In contrast to most existing distributed PHD filters, our filter employs an `arithmetic average' fusion. For particles--GM conversion, we use a method that avoids particle clustering and enables a significance-based pruning of the GM components. For GM--particles conversion, we develop an importance sampling based method that enables a parallelization of filtering and dissemination/fusion operations. The proposed distributed particle-PHD filter is able to integrate GM-based local PHD filters. Simulations demonstrate the excellent performance and small communication and computation requirements of our filter.