Fault diagnosis in robot swarms: An adaptive online behaviour characterisation approach

James O'Keeffe; Danesh Tarapore; Alan G. Millard; Jon Timmis

doi:10.1109/SSCI.2017.8280891

Fault diagnosis in robot swarms: An adaptive online behaviour characterisation approach

James O'Keeffe^*, Danesh Tarapore, Alan G. Millard, Jon Timmis

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The need for an active approach to fault tolerance in swarm robotics systems is well established. This will necessarily include an approach to fault diagnosis if robot swarms are to retain long-term autonomy. This paper proposes a novel method for fault diagnosis, based around behavioural feature vectors, that incorporates real-time learning and memory. Initial results are encouraging, and show that an unsupervised learning approach is able to diagnose common electro-mechanical fault types, and arrive at an appropriate recovery option in the majority of the cases tested.

Original language	English
Title of host publication	2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings
Publisher	IEEE
Pages	1-8
Number of pages	8
ISBN (Electronic)	9781538627259
DOIs	https://doi.org/10.1109/SSCI.2017.8280891
Publication status	Published - 5 Feb 2018
Event	2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Honolulu, United States Duration: 27 Nov 2017 → 1 Dec 2017

Conference

Conference	2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017
Country/Territory	United States
City	Honolulu
Period	27/11/17 → 1/12/17

Access to Document

10.1109/SSCI.2017.8280891

Cite this

@inproceedings{eff2da5d72ca43ea863dda5de3513b0a,

title = "Fault diagnosis in robot swarms: An adaptive online behaviour characterisation approach",

abstract = "The need for an active approach to fault tolerance in swarm robotics systems is well established. This will necessarily include an approach to fault diagnosis if robot swarms are to retain long-term autonomy. This paper proposes a novel method for fault diagnosis, based around behavioural feature vectors, that incorporates real-time learning and memory. Initial results are encouraging, and show that an unsupervised learning approach is able to diagnose common electro-mechanical fault types, and arrive at an appropriate recovery option in the majority of the cases tested.",

author = "James O'Keeffe and Danesh Tarapore and Millard, {Alan G.} and Jon Timmis",

year = "2018",

month = feb,

day = "5",

doi = "10.1109/SSCI.2017.8280891",

language = "English",

pages = "1--8",

booktitle = "2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings",

publisher = "IEEE",

address = "United States",

note = "2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 ; Conference date: 27-11-2017 Through 01-12-2017",

}

O'Keeffe, J , Tarapore, D , Millard, AG & Timmis, J 2018, Fault diagnosis in robot swarms: An adaptive online behaviour characterisation approach. in 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings. IEEE, pp. 1-8, 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017, Honolulu, United States, 27/11/17. https://doi.org/10.1109/SSCI.2017.8280891

TY - GEN

T1 - Fault diagnosis in robot swarms

T2 - 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017

AU - O'Keeffe, James

AU - Tarapore, Danesh

AU - Millard, Alan G.

AU - Timmis, Jon

PY - 2018/2/5

Y1 - 2018/2/5

N2 - The need for an active approach to fault tolerance in swarm robotics systems is well established. This will necessarily include an approach to fault diagnosis if robot swarms are to retain long-term autonomy. This paper proposes a novel method for fault diagnosis, based around behavioural feature vectors, that incorporates real-time learning and memory. Initial results are encouraging, and show that an unsupervised learning approach is able to diagnose common electro-mechanical fault types, and arrive at an appropriate recovery option in the majority of the cases tested.

AB - The need for an active approach to fault tolerance in swarm robotics systems is well established. This will necessarily include an approach to fault diagnosis if robot swarms are to retain long-term autonomy. This paper proposes a novel method for fault diagnosis, based around behavioural feature vectors, that incorporates real-time learning and memory. Initial results are encouraging, and show that an unsupervised learning approach is able to diagnose common electro-mechanical fault types, and arrive at an appropriate recovery option in the majority of the cases tested.

UR - http://www.scopus.com/inward/record.url?scp=85046072955&partnerID=8YFLogxK

U2 - 10.1109/SSCI.2017.8280891

DO - 10.1109/SSCI.2017.8280891

M3 - Conference contribution

AN - SCOPUS:85046072955

SP - 1

EP - 8

BT - 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings

PB - IEEE

Y2 - 27 November 2017 through 1 December 2017

ER -

Fault diagnosis in robot swarms: An adaptive online behaviour characterisation approach

Abstract

Conference

Access to Document

Other files and links

Cite this