Deterministic sampling-based switching Kalman filtering for vehicle tracking

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Accurate tracking is a difficult task in most computer vision applications. Errors in target localization and tracking result not only from the general uncontrolled nature of the environment, but also from inaccurate modeling of the target motion. This work presents a novel solution for the robust estimation of target trajectories obtained from real-world scenes such as traffic intersections. The main contribution of this work is a deterministic sampling approach applied to the filtering step of the switching Kalman filter/smoother. The unscented transform is used to obtain a fixed set of samples of the state distribution in the filtering step. Results demonstrating the improved accuracy and robustness of the proposed method, namely, deterministic sampling or unscented transform-based switching Kalman filter (DS-SKS or UKS) and the standard switching Kalman filter/smoother (SKS) are presented.

Original languageEnglish (US)
Title of host publicationProceedings of ITSC 2006
Subtitle of host publication2006 IEEE Intelligent Transportation Systems Conference
Pages1340-1345
Number of pages6
StatePublished - Dec 1 2006
EventITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference - Toronto, ON, Canada
Duration: Sep 17 2006Sep 20 2006

Publication series

NameIEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC

Other

OtherITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference
CountryCanada
CityToronto, ON
Period9/17/069/20/06

Fingerprint

Kalman filters
Sampling
Computer vision
Trajectories
Mathematical transformations

Keywords

  • Deterministic sampling
  • Smoothing
  • Switching Kalman filtering
  • Unscented transform

Cite this

Veeraraghavan, H., Papanikolopoulos, N. P., & Schrater, P. R. (2006). Deterministic sampling-based switching Kalman filtering for vehicle tracking. In Proceedings of ITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference (pp. 1340-1345). [1707409] (IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC).

Deterministic sampling-based switching Kalman filtering for vehicle tracking. / Veeraraghavan, Harini; Papanikolopoulos, Nikolaos P; Schrater, Paul R.

Proceedings of ITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference. 2006. p. 1340-1345 1707409 (IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Veeraraghavan, H, Papanikolopoulos, NP & Schrater, PR 2006, Deterministic sampling-based switching Kalman filtering for vehicle tracking. in Proceedings of ITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference., 1707409, IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC, pp. 1340-1345, ITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference, Toronto, ON, Canada, 9/17/06.
Veeraraghavan H, Papanikolopoulos NP, Schrater PR. Deterministic sampling-based switching Kalman filtering for vehicle tracking. In Proceedings of ITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference. 2006. p. 1340-1345. 1707409. (IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC).
Veeraraghavan, Harini ; Papanikolopoulos, Nikolaos P ; Schrater, Paul R. / Deterministic sampling-based switching Kalman filtering for vehicle tracking. Proceedings of ITSC 2006: 2006 IEEE Intelligent Transportation Systems Conference. 2006. pp. 1340-1345 (IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC).
@inproceedings{0681d6e631344282b2f5d61e732ff412,
title = "Deterministic sampling-based switching Kalman filtering for vehicle tracking",
abstract = "Accurate tracking is a difficult task in most computer vision applications. Errors in target localization and tracking result not only from the general uncontrolled nature of the environment, but also from inaccurate modeling of the target motion. This work presents a novel solution for the robust estimation of target trajectories obtained from real-world scenes such as traffic intersections. The main contribution of this work is a deterministic sampling approach applied to the filtering step of the switching Kalman filter/smoother. The unscented transform is used to obtain a fixed set of samples of the state distribution in the filtering step. Results demonstrating the improved accuracy and robustness of the proposed method, namely, deterministic sampling or unscented transform-based switching Kalman filter (DS-SKS or UKS) and the standard switching Kalman filter/smoother (SKS) are presented.",
keywords = "Deterministic sampling, Smoothing, Switching Kalman filtering, Unscented transform",
author = "Harini Veeraraghavan and Papanikolopoulos, {Nikolaos P} and Schrater, {Paul R}",
year = "2006",
month = "12",
day = "1",
language = "English (US)",
isbn = "1424400945",
series = "IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC",
pages = "1340--1345",
booktitle = "Proceedings of ITSC 2006",

}

TY - GEN

T1 - Deterministic sampling-based switching Kalman filtering for vehicle tracking

AU - Veeraraghavan, Harini

AU - Papanikolopoulos, Nikolaos P

AU - Schrater, Paul R

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Accurate tracking is a difficult task in most computer vision applications. Errors in target localization and tracking result not only from the general uncontrolled nature of the environment, but also from inaccurate modeling of the target motion. This work presents a novel solution for the robust estimation of target trajectories obtained from real-world scenes such as traffic intersections. The main contribution of this work is a deterministic sampling approach applied to the filtering step of the switching Kalman filter/smoother. The unscented transform is used to obtain a fixed set of samples of the state distribution in the filtering step. Results demonstrating the improved accuracy and robustness of the proposed method, namely, deterministic sampling or unscented transform-based switching Kalman filter (DS-SKS or UKS) and the standard switching Kalman filter/smoother (SKS) are presented.

AB - Accurate tracking is a difficult task in most computer vision applications. Errors in target localization and tracking result not only from the general uncontrolled nature of the environment, but also from inaccurate modeling of the target motion. This work presents a novel solution for the robust estimation of target trajectories obtained from real-world scenes such as traffic intersections. The main contribution of this work is a deterministic sampling approach applied to the filtering step of the switching Kalman filter/smoother. The unscented transform is used to obtain a fixed set of samples of the state distribution in the filtering step. Results demonstrating the improved accuracy and robustness of the proposed method, namely, deterministic sampling or unscented transform-based switching Kalman filter (DS-SKS or UKS) and the standard switching Kalman filter/smoother (SKS) are presented.

KW - Deterministic sampling

KW - Smoothing

KW - Switching Kalman filtering

KW - Unscented transform

UR - http://www.scopus.com/inward/record.url?scp=41849144917&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=41849144917&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1424400945

SN - 9781424400942

T3 - IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC

SP - 1340

EP - 1345

BT - Proceedings of ITSC 2006

ER -