Merging-based forward-backward smoothing on Gaussian mixtures

Author

Rahmathullah, Abu Sajana ; Svensson, Lars ; Svensson, Daniel

Author_Institution

Dept. of Signals & Syst., Chalmers Univ. of Technol., Gothenburg, Sweden

fYear

2014

fDate

7-10 July 2014

Firstpage

1

Lastpage

8

Abstract

Conventional forward-backward smoothing (FBS) for Gaussian mixture (GM) problems are based on pruning methods which yield a degenerate hypothesis tree and often lead to underestimated uncertainties. To overcome these shortcomings, we propose an algorithm that is based on merging components in the GM during filtering and smoothing. Compared to FBS based on the N-scan pruning, the proposed algorithm offers better performance in terms of track loss, root mean squared error (RMSE) and normalized estimation error squared (NEES) without increasing the computational complexity.

Keywords

Gaussian processes; computational complexity; least mean squares methods; merging; mixture models; smoothing methods; FBS; GM; Gaussian mixtures; N-scan pruning; NEES; RMSE; computational complexity; filtering; merging-based forward-backward smoothing; normalized estimation error square; root mean squared error; track loss; Approximation methods; Clutter; Hafnium; Merging; Smoothing methods; Target tracking; Uncertainty; Gaussian mixtures; data association; filtering; forward-backward smoothing; smoothing;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Fusion (FUSION), 2014 17th International Conference on

Conference_Location

Salamanca

Type

conf

Filename

6916248