Multitarget state and track estimation for the probability hypotheses density filter

Abstract

The particle Probability Hypotheses Density (particle-PHD) filter is a tractable approach for Random Finite Set (RFS) Bayes estimation, but the particle-PHD filter can not directly derive the target track. Most existing approaches combine the data association step to solve this problem. This paper proposes an algorithm which does not need the association step. Our basic ideal is based on the clustering algorithm of Finite Mixture Models (FMM). The intensity distribution is first derived by the particle-PHD filter, and then the clustering algorithm is applied to estimate the multitarget states and tracks jointly. The clustering process includes two steps: the prediction and update. The key to the proposed algorithm is to use the prediction as the initial points and the convergent points as the estimates. Besides, Expectation-Maximization (EM) and Markov Chain Monte Carlo (MCMC) approaches are used for the FMM parameter estimation.

State Estimation and Smoothing for the Probability Hypothesis Density Filter

“…In the PHD filter case, each component is usually modelled as a multivariate Gaussian distribution θ j k = N ( xj k , Σ j k ). State estimation using the Gibbs sampler was also considered in [63], where the authors compared estimation using the EM algorithm and Bayesian estimation with the Gibbs sampler. However, the authors proposed a complete mixture modeling estimation procedure and results were validated in a simple toy example.…”

State Estimation and Smoothing for the Probability Hypothesis Density Filter

“…Parametric estimation for the PHD filter using the EM algorithm has been discussed in the literature for the standard SNR case [15]. More recently, Bayesian state estimation using MCMC was proposed in [63]. In Section 3.6, a new MCMC method for state estimation is provided, which takes advantage of the noise conditions commonly found in visual tracking applications.…”