Recognising agent behaviour during variable length activities

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

Abstract

In this paper we present a new method for obtaining situation awareness via the automatic recognition of agent behaviours. In contrast to many other approaches, the presented method models different behaviour durations without using a fixed classification window, and does not require a distribution of behaviour durations. We introduce the Variable Window Layered Hidden Markov Model (VW-LHMM) as an extension of the LHMM to specifically address behaviours with irregular duration. We validate our approach by simulating three high-level behaviours within the harbour and coastline security domain. We compare performance against the LHMM and show that our approach provides a 10% improvement in classification accuracy, in addition to earlier classification. © 2010 The authors and IOS Press. All rights reserved.

Original languageEnglish
Title of host publicationECAI 2010
Pages803-808
Number of pages6
Volume215
DOIs
Publication statusPublished - 2010

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume215
ISSN (Print)0922-6389

Fingerprint

Hidden Markov models
Ports and harbors

Cite this

Baxter, R., Lane, D., & Petillot, Y. (2010). Recognising agent behaviour during variable length activities. In ECAI 2010 (Vol. 215, pp. 803-808). (Frontiers in Artificial Intelligence and Applications; Vol. 215). https://doi.org/10.3233/978-1-60750-606-5-803
Baxter, Rolf ; Lane, David ; Petillot, Yvan. / Recognising agent behaviour during variable length activities. ECAI 2010. Vol. 215 2010. pp. 803-808 (Frontiers in Artificial Intelligence and Applications).
@inproceedings{91de3c34967a43f4b73b47bc91e9e1ab,
title = "Recognising agent behaviour during variable length activities",
abstract = "In this paper we present a new method for obtaining situation awareness via the automatic recognition of agent behaviours. In contrast to many other approaches, the presented method models different behaviour durations without using a fixed classification window, and does not require a distribution of behaviour durations. We introduce the Variable Window Layered Hidden Markov Model (VW-LHMM) as an extension of the LHMM to specifically address behaviours with irregular duration. We validate our approach by simulating three high-level behaviours within the harbour and coastline security domain. We compare performance against the LHMM and show that our approach provides a 10{\%} improvement in classification accuracy, in addition to earlier classification. {\circledC} 2010 The authors and IOS Press. All rights reserved.",
author = "Rolf Baxter and David Lane and Yvan Petillot",
year = "2010",
doi = "10.3233/978-1-60750-606-5-803",
language = "English",
isbn = "9781607506058",
volume = "215",
series = "Frontiers in Artificial Intelligence and Applications",
pages = "803--808",
booktitle = "ECAI 2010",

}

Baxter, R, Lane, D & Petillot, Y 2010, Recognising agent behaviour during variable length activities. in ECAI 2010. vol. 215, Frontiers in Artificial Intelligence and Applications, vol. 215, pp. 803-808. https://doi.org/10.3233/978-1-60750-606-5-803

Recognising agent behaviour during variable length activities. / Baxter, Rolf; Lane, David; Petillot, Yvan.

ECAI 2010. Vol. 215 2010. p. 803-808 (Frontiers in Artificial Intelligence and Applications; Vol. 215).

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

TY - GEN

T1 - Recognising agent behaviour during variable length activities

AU - Baxter, Rolf

AU - Lane, David

AU - Petillot, Yvan

PY - 2010

Y1 - 2010

N2 - In this paper we present a new method for obtaining situation awareness via the automatic recognition of agent behaviours. In contrast to many other approaches, the presented method models different behaviour durations without using a fixed classification window, and does not require a distribution of behaviour durations. We introduce the Variable Window Layered Hidden Markov Model (VW-LHMM) as an extension of the LHMM to specifically address behaviours with irregular duration. We validate our approach by simulating three high-level behaviours within the harbour and coastline security domain. We compare performance against the LHMM and show that our approach provides a 10% improvement in classification accuracy, in addition to earlier classification. © 2010 The authors and IOS Press. All rights reserved.

AB - In this paper we present a new method for obtaining situation awareness via the automatic recognition of agent behaviours. In contrast to many other approaches, the presented method models different behaviour durations without using a fixed classification window, and does not require a distribution of behaviour durations. We introduce the Variable Window Layered Hidden Markov Model (VW-LHMM) as an extension of the LHMM to specifically address behaviours with irregular duration. We validate our approach by simulating three high-level behaviours within the harbour and coastline security domain. We compare performance against the LHMM and show that our approach provides a 10% improvement in classification accuracy, in addition to earlier classification. © 2010 The authors and IOS Press. All rights reserved.

UR - http://www.scopus.com/inward/record.url?scp=77956040405&partnerID=8YFLogxK

U2 - 10.3233/978-1-60750-606-5-803

DO - 10.3233/978-1-60750-606-5-803

M3 - Conference contribution

SN - 9781607506058

VL - 215

T3 - Frontiers in Artificial Intelligence and Applications

SP - 803

EP - 808

BT - ECAI 2010

ER -

Baxter R, Lane D, Petillot Y. Recognising agent behaviour during variable length activities. In ECAI 2010. Vol. 215. 2010. p. 803-808. (Frontiers in Artificial Intelligence and Applications). https://doi.org/10.3233/978-1-60750-606-5-803