Bayesian tracking and parameter learning for non-linear multiple target tracking models

TL;DR

This paper introduces a Bayesian MCMC-based algorithm for non-linear, non-Gaussian multiple target tracking that jointly estimates target states, associations, and model parameters, showing superior performance over existing methods.

Contribution

The paper presents a novel Bayesian MCMC algorithm for joint target tracking and parameter learning in complex non-linear, non-Gaussian MTT models, improving accuracy and robustness.

Findings

Significant performance improvements over competing techniques

Effective joint estimation of target states and model parameters

Robust tracking in non-linear, non-Gaussian scenarios

Abstract

We propose a new Bayesian tracking and parameter learning algorithm for non-linear non-Gaussian multiple target tracking (MTT) models. We design a Markov chain Monte Carlo (MCMC) algorithm to sample from the posterior distribution of the target states, birth and death times, and association of observations to targets, which constitutes the solution to the tracking problem, as well as the model parameters. In the numerical section, we present performance comparisons with several competing techniques and demonstrate significant performance improvements in all cases.