An alternative derivation of a Bayes tracking filter based on finite mixture models

Author

Liu, Weifeng ; Han, Chongzhao ; Lian, Feng

Author_Institution

Electron. Inf. Engr., Xi´´an Jiaotong Univ., Xi´´an, China

fYear

2009

fDate

6-9 July 2009

Firstpage

842

Lastpage

849

Abstract

Ba-Tuong-Vo et al proposed a Bayes filter of single target in the random finite set framework. In this paper, we first extend the parameter mixture models (PMM) to state mixture models(s). And further an alternative derivation of a Bayesian tracking filter in clutter is proposed for single target. The key of the proposed algorithm is to derive the measurement likelihood function based on finite mixture models. In addition, a closed-form recursion under the linear Gaussian assumption is discussed.

Keywords

Bayes methods; Gaussian processes; filtering theory; maximum likelihood estimation; recursive estimation; target tracking; Bayes tracking filter; closed-form recursion assumption; finite mixture models; linear Gaussian assumption; measurement likelihood function; parameter mixture models; random finite set framework; state mixture models; Bayesian methods; Equations; Information filtering; Information filters; Measurement uncertainty; Object detection; State estimation; Target tracking; Bayes estimation; Single target tracking; finite mixture models; likelihood;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Fusion, 2009. FUSION '09. 12th International Conference on

Conference_Location

Seattle, WA

Print_ISBN

978-0-9824-4380-4

Type

conf

Filename

5203690