Fault diagnosis in robot swarms: an adaptive online behaviour characterisation approach
Fault diagnosis in robot swarms: an adaptive online behaviour characterisation approach
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.
1-8
Institute of Electrical and Electronics Engineers Inc.
O'Keeffe, James
51c12624-5488-4535-9dcf-a57c5175659e
Tarapore, Danesh
fe8ec8ae-1fad-4726-abef-84b538542ee4
Millard, Alan G.
58064324-77bf-4ec7-ba6d-7c2763e02954
Timmis, Jon
b68f4b8e-6192-4caf-858d-8185f6e7c66f
2017
O'Keeffe, James
51c12624-5488-4535-9dcf-a57c5175659e
Tarapore, Danesh
fe8ec8ae-1fad-4726-abef-84b538542ee4
Millard, Alan G.
58064324-77bf-4ec7-ba6d-7c2763e02954
Timmis, Jon
b68f4b8e-6192-4caf-858d-8185f6e7c66f
O'Keeffe, James, Tarapore, Danesh, Millard, Alan G. and Timmis, Jon
(2017)
Fault diagnosis in robot swarms: an adaptive online behaviour characterisation approach.
In 2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings.
vol. 2018-January,
Institute of Electrical and Electronics Engineers Inc.
.
(doi:10.1109/SSCI.2017.8280891).
Record type:
Conference or Workshop Item
(Paper)
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.
Full text not available from this repository.
More information
Published date: 2017
Venue - Dates:
2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017, Honolulu, United States, 2017-11-27 - 2017-12-01
Identifiers
Local EPrints ID: 420616
URI: https://eprints.soton.ac.uk/id/eprint/420616
PURE UUID: 5e36b783-95d4-4941-b023-77c01819f706
Catalogue record
Date deposited: 11 May 2018 16:30
Last modified: 12 Nov 2019 01:28
Export record
Altmetrics
Contributors
Author:
James O'Keeffe
Author:
Alan G. Millard
Author:
Jon Timmis
University divisions
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
