Interactive Model Fusion-Based GM-PHD Filter

TL;DR

This paper introduces an interactive fusion method that transforms non-Gaussian measurement noise in multi-target tracking into a Gaussian framework, enabling effective state estimation through a multi-model GM-PHD filter approach.

Contribution

It develops a novel transformation of non-Gaussian noise into Gaussian mixtures and employs an interactive multi-model fusion to improve multi-target tracking accuracy.

Findings

Enhanced tracking performance under non-Gaussian noise conditions

Effective multi-model fusion improves state estimation accuracy

Simulation results validate the approach's effectiveness

Abstract

In multi-target tracking (MTT), non-Gaussian measurement noise from sensors can diminish the performance of the Gaussian-assumed Gaussian mixture probability hypothesis density (GM-PHD) filter. In this paper, an approach that transforms the MTT problem under non-Gaussian conditions into an MTT problem under Gaussian conditions is developed. Specifically, measurement noise with a non-Gaussian distribution is modeled as a weighted sum of different Gaussian distributions. Subsequently, the GM-PHD filter is applied to compute the multi-target states under these distinct Gaussian distributions. Finally, an interactive multi-model framework is employed to fuse the diverse multi-target state information into a unified synthesis. The effectiveness of the proposed approach is validated through the simulation results.