A Bayesian analysis of the mixed labelling phenomenon in two-target tracking

E.H. Aoki, Y. Boers, L. Svensson, Pranab K. Mandal, Arunabha Bagchi

Research output: Book/ReportReport

Abstract

In mulit-target tracking and labelling (MTTL), mixed labelling corresponds to a situation where there is ambiguity in labelling, i.e. in the assignment of labels to locations (where a "location" here means simply an unlabelled single-target state. The phenomenon is well-known in literature, and known to occur in the situation where targets move in close proximity to each other and afterwards separate. The occurrence of mixed labelling has been empirically observed using particle filter implementations of the Bayesian MTTL recursion. In this memorandum, we will instead demonstrate the occurrence of mixed labelling (in the situation of closely spaced targets) using only the Bayesian recursion itself, for a scenario containing two targets and no target births or deaths. We will also show how mixed labelling generally persists after the targets become well-separated, and how mixed labelling might not happen when the unlabelled single-target state contains non-kinematic quantities.
LanguageUndefined
Place of PublicationEnschede
PublisherDepartment of Applied Mathematics, University of Twente
Number of pages4
StatePublished - Aug 2012

Publication series

Name
No.1986
ISSN (Print)1874-4850
ISSN (Electronic)1874-4850

Keywords

  • METIS-296072
  • EWI-22144

Cite this

Aoki, E. H., Boers, Y., Svensson, L., Mandal, P. K., & Bagchi, A. (2012). A Bayesian analysis of the mixed labelling phenomenon in two-target tracking. Enschede: Department of Applied Mathematics, University of Twente.
Aoki, E.H. ; Boers, Y. ; Svensson, L. ; Mandal, Pranab K. ; Bagchi, Arunabha. / A Bayesian analysis of the mixed labelling phenomenon in two-target tracking. Enschede : Department of Applied Mathematics, University of Twente, 2012. 4 p.
@book{ea60abddb2bf490fb16f0b3526a4dc5a,
title = "A Bayesian analysis of the mixed labelling phenomenon in two-target tracking",
abstract = "In mulit-target tracking and labelling (MTTL), mixed labelling corresponds to a situation where there is ambiguity in labelling, i.e. in the assignment of labels to locations (where a {"}location{"} here means simply an unlabelled single-target state. The phenomenon is well-known in literature, and known to occur in the situation where targets move in close proximity to each other and afterwards separate. The occurrence of mixed labelling has been empirically observed using particle filter implementations of the Bayesian MTTL recursion. In this memorandum, we will instead demonstrate the occurrence of mixed labelling (in the situation of closely spaced targets) using only the Bayesian recursion itself, for a scenario containing two targets and no target births or deaths. We will also show how mixed labelling generally persists after the targets become well-separated, and how mixed labelling might not happen when the unlabelled single-target state contains non-kinematic quantities.",
keywords = "METIS-296072, EWI-22144",
author = "E.H. Aoki and Y. Boers and L. Svensson and Mandal, {Pranab K.} and Arunabha Bagchi",
year = "2012",
month = "8",
language = "Undefined",
publisher = "Department of Applied Mathematics, University of Twente",
number = "1986",

}

Aoki, EH, Boers, Y, Svensson, L, Mandal, PK & Bagchi, A 2012, A Bayesian analysis of the mixed labelling phenomenon in two-target tracking. Department of Applied Mathematics, University of Twente, Enschede.

A Bayesian analysis of the mixed labelling phenomenon in two-target tracking. / Aoki, E.H.; Boers, Y.; Svensson, L.; Mandal, Pranab K.; Bagchi, Arunabha.

Enschede : Department of Applied Mathematics, University of Twente, 2012. 4 p.

Research output: Book/ReportReport

TY - BOOK

T1 - A Bayesian analysis of the mixed labelling phenomenon in two-target tracking

AU - Aoki,E.H.

AU - Boers,Y.

AU - Svensson,L.

AU - Mandal,Pranab K.

AU - Bagchi,Arunabha

PY - 2012/8

Y1 - 2012/8

N2 - In mulit-target tracking and labelling (MTTL), mixed labelling corresponds to a situation where there is ambiguity in labelling, i.e. in the assignment of labels to locations (where a "location" here means simply an unlabelled single-target state. The phenomenon is well-known in literature, and known to occur in the situation where targets move in close proximity to each other and afterwards separate. The occurrence of mixed labelling has been empirically observed using particle filter implementations of the Bayesian MTTL recursion. In this memorandum, we will instead demonstrate the occurrence of mixed labelling (in the situation of closely spaced targets) using only the Bayesian recursion itself, for a scenario containing two targets and no target births or deaths. We will also show how mixed labelling generally persists after the targets become well-separated, and how mixed labelling might not happen when the unlabelled single-target state contains non-kinematic quantities.

AB - In mulit-target tracking and labelling (MTTL), mixed labelling corresponds to a situation where there is ambiguity in labelling, i.e. in the assignment of labels to locations (where a "location" here means simply an unlabelled single-target state. The phenomenon is well-known in literature, and known to occur in the situation where targets move in close proximity to each other and afterwards separate. The occurrence of mixed labelling has been empirically observed using particle filter implementations of the Bayesian MTTL recursion. In this memorandum, we will instead demonstrate the occurrence of mixed labelling (in the situation of closely spaced targets) using only the Bayesian recursion itself, for a scenario containing two targets and no target births or deaths. We will also show how mixed labelling generally persists after the targets become well-separated, and how mixed labelling might not happen when the unlabelled single-target state contains non-kinematic quantities.

KW - METIS-296072

KW - EWI-22144

M3 - Report

BT - A Bayesian analysis of the mixed labelling phenomenon in two-target tracking

PB - Department of Applied Mathematics, University of Twente

CY - Enschede

ER -

Aoki EH, Boers Y, Svensson L, Mandal PK, Bagchi A. A Bayesian analysis of the mixed labelling phenomenon in two-target tracking. Enschede: Department of Applied Mathematics, University of Twente, 2012. 4 p.