IMM Bernoulli Gaussian Particle Filter

Olivér Törő, Tamás Bécsi, Szilárd Aradi, P. Gáspár

Research output: Contribution to journalArticle

Abstract

The Bernoulli filter (BF) in the interacting multiple model (IMM) framework is proposed for detecting and tracking a maneuvering target. The BF is implemented as a particle filter and embedded in the IMM structure. The communication between the IMM and the BF is achieved through a Gaussian layer. Particles are drawn from the mixture densities at each recursion and the model likelihoods are computed from the filter innovations. Simulations show that the proposed filter outperforms the single model variant and it can effectively choose the correct motion model and estimate the state of the tracked object.

Original languageEnglish
Pages (from-to)274-279
Number of pages6
JournalIFAC-PapersOnLine
Volume51
Issue number22
DOIs
Publication statusPublished - Jan 1 2018

Fingerprint

Model structures
Innovation
Communication

Keywords

  • Bernoulli filter
  • estimation
  • multiple-model
  • object tracking
  • particle filter

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

IMM Bernoulli Gaussian Particle Filter. / Törő, Olivér; Bécsi, Tamás; Aradi, Szilárd; Gáspár, P.

In: IFAC-PapersOnLine, Vol. 51, No. 22, 01.01.2018, p. 274-279.

Research output: Contribution to journalArticle

Törő, Olivér ; Bécsi, Tamás ; Aradi, Szilárd ; Gáspár, P. / IMM Bernoulli Gaussian Particle Filter. In: IFAC-PapersOnLine. 2018 ; Vol. 51, No. 22. pp. 274-279.
@article{06812da93dec48b0952060f021a08730,
title = "IMM Bernoulli Gaussian Particle Filter",
abstract = "The Bernoulli filter (BF) in the interacting multiple model (IMM) framework is proposed for detecting and tracking a maneuvering target. The BF is implemented as a particle filter and embedded in the IMM structure. The communication between the IMM and the BF is achieved through a Gaussian layer. Particles are drawn from the mixture densities at each recursion and the model likelihoods are computed from the filter innovations. Simulations show that the proposed filter outperforms the single model variant and it can effectively choose the correct motion model and estimate the state of the tracked object.",
keywords = "Bernoulli filter, estimation, multiple-model, object tracking, particle filter",
author = "Oliv{\'e}r T{\"o}rő and Tam{\'a}s B{\'e}csi and Szil{\'a}rd Aradi and P. G{\'a}sp{\'a}r",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.ifacol.2018.11.554",
language = "English",
volume = "51",
pages = "274--279",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "IFAC Secretariat",
number = "22",

}

TY - JOUR

T1 - IMM Bernoulli Gaussian Particle Filter

AU - Törő, Olivér

AU - Bécsi, Tamás

AU - Aradi, Szilárd

AU - Gáspár, P.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The Bernoulli filter (BF) in the interacting multiple model (IMM) framework is proposed for detecting and tracking a maneuvering target. The BF is implemented as a particle filter and embedded in the IMM structure. The communication between the IMM and the BF is achieved through a Gaussian layer. Particles are drawn from the mixture densities at each recursion and the model likelihoods are computed from the filter innovations. Simulations show that the proposed filter outperforms the single model variant and it can effectively choose the correct motion model and estimate the state of the tracked object.

AB - The Bernoulli filter (BF) in the interacting multiple model (IMM) framework is proposed for detecting and tracking a maneuvering target. The BF is implemented as a particle filter and embedded in the IMM structure. The communication between the IMM and the BF is achieved through a Gaussian layer. Particles are drawn from the mixture densities at each recursion and the model likelihoods are computed from the filter innovations. Simulations show that the proposed filter outperforms the single model variant and it can effectively choose the correct motion model and estimate the state of the tracked object.

KW - Bernoulli filter

KW - estimation

KW - multiple-model

KW - object tracking

KW - particle filter

UR - http://www.scopus.com/inward/record.url?scp=85058158816&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058158816&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2018.11.554

DO - 10.1016/j.ifacol.2018.11.554

M3 - Article

AN - SCOPUS:85058158816

VL - 51

SP - 274

EP - 279

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 22

ER -