Particle based MAP state estimation: A comparison

Author

Saha, S. ; Boers, Y. ; Driessen, H. ; Mandal, P.K. ; Bagchi, A.

Author_Institution

Dept. of Appl. Math., Univ. of Twente, Enschede, Netherlands

fYear

2009

fDate

6-9 July 2009

Firstpage

278

Lastpage

283

Abstract

MAP estimation is a good alternative to MMSE for certain applications involving nonlinear non Gaussian systems. Recently a new particle filter based MAP estimator has been derived. This new method extracts the MAP directly from the output of a running particle filter. In the recent past, a Viterbi algorithm based MAP sequence estimator has been developed. In this paper, we compare these two methods for estimating the current state and the numerical results show that the former performs better.

Keywords

Monte Carlo methods; least mean squares methods; maximum likelihood estimation; nonlinear filters; particle filtering (numerical methods); MAP sequence; MMSE; Monte Carlo method; Viterbi algorithm; maximum a posteriori estimate; minimum mean square error method; nonlinear non Gaussian system; particle filter based MAP state estimation; Aircraft navigation; Airplanes; Bayesian methods; Clouds; Density measurement; Mathematics; Monte Carlo methods; Particle filters; State estimation; Viterbi algorithm; Bayesian point estimation; filter MAP; particle filter; sequential Monte Carlo;

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

5203651