Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling

James C. Ward; Ryan McConville; Edmund R. Hunt

doi:10.1145/3672608.3707752

Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling

James C. Ward
, Ryan McConville
, Edmund R. Hunt

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

3 Citations (Scopus)

66 Downloads (Pure)

Abstract

The problem of decentralized multi-robot patrol has previously been approached primarily with hand-designed strategies for minimization of "idlenes" over the vertices of a graph-structured environment. Here we present two lightweight neural network-based strategies to tackle this problem, and show that they significantly outperform existing strategies in both idleness minimization and against an intelligent intruder model, as well as presenting an examination of robustness to communication failure. Our results also indicate important considerations for future strategy design.

Original language	English
Title of host publication	SAC '25
Subtitle of host publication	Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing
Publisher	Association for Computing Machinery
Pages	823-831
Number of pages	9
ISBN (Electronic)	9798400706295
DOIs	https://doi.org/10.1145/3672608.3707752
Publication status	Published - 14 May 2025
Event	40th ACM/SIGAPP Symposium on Applied Computing - Sicily, Italy Duration: 31 Mar 2025 → 4 Apr 2025 Conference number: 40th https://www.sigapp.org/sac/sac2025/

Publication series

Name	Proceedings of the ACM/SIGAPP Symposium on Applied Computing
Publisher	ACM

Conference

Conference	40th ACM/SIGAPP Symposium on Applied Computing
Abbreviated title	SAC '25
Country/Territory	Italy
City	Sicily
Period	31/03/25 → 4/04/25
Internet address	https://www.sigapp.org/sac/sac2025/

Bibliographical note

Publisher Copyright:
© 2025 Copyright held by the owner/author(s).

Keywords

cs.RO

Access to Document

10.1145/3672608.3707752Licence: CC BY

Full-text PDF (pre-print)Submitted manuscript, 783 KBLicence: CC BY

Handle.net

Persistent link

Cite this

Ward, J. C., McConville, R., & Hunt, E. R. (2025). Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling. In SAC '25: Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing (pp. 823-831). (Proceedings of the ACM/SIGAPP Symposium on Applied Computing). Association for Computing Machinery. https://doi.org/10.1145/3672608.3707752

@inproceedings{3f147cb84cae4e3ab1724e72a4ce6941,

title = "Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling",

abstract = "The problem of decentralized multi-robot patrol has previously been approached primarily with hand-designed strategies for minimization of {"}idlenes{"} over the vertices of a graph-structured environment. Here we present two lightweight neural network-based strategies to tackle this problem, and show that they significantly outperform existing strategies in both idleness minimization and against an intelligent intruder model, as well as presenting an examination of robustness to communication failure. Our results also indicate important considerations for future strategy design.",

keywords = "cs.RO",

author = "Ward, \{James C.\} and Ryan McConville and Hunt, \{Edmund R.\}",

note = "Publisher Copyright: {\textcopyright} 2025 Copyright held by the owner/author(s).; 40th ACM/SIGAPP Symposium on Applied Computing, SAC '25 ; Conference date: 31-03-2025 Through 04-04-2025",

year = "2025",

month = may,

day = "14",

doi = "10.1145/3672608.3707752",

language = "English",

series = "Proceedings of the ACM/SIGAPP Symposium on Applied Computing",

publisher = "Association for Computing Machinery",

pages = "823--831",

booktitle = "SAC '25",

address = "United States",

url = "https://www.sigapp.org/sac/sac2025/",

}

Ward, JC, McConville, R & Hunt, ER 2025, Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling. in SAC '25: Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing. Proceedings of the ACM/SIGAPP Symposium on Applied Computing, Association for Computing Machinery, pp. 823-831, 40th ACM/SIGAPP Symposium on Applied Computing, Sicily, Italy, 31/03/25. https://doi.org/10.1145/3672608.3707752

Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling. / Ward, James C.; McConville, Ryan; Hunt, Edmund R.
SAC '25: Proceedings of the 40th ACM/SIGAPP Symposium on Applied Computing. Association for Computing Machinery, 2025. p. 823-831 (Proceedings of the ACM/SIGAPP Symposium on Applied Computing).

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling

AU - Ward, James C.

AU - McConville, Ryan

AU - Hunt, Edmund R.

N1 - Conference code: 40th

PY - 2025/5/14

Y1 - 2025/5/14

N2 - The problem of decentralized multi-robot patrol has previously been approached primarily with hand-designed strategies for minimization of "idlenes" over the vertices of a graph-structured environment. Here we present two lightweight neural network-based strategies to tackle this problem, and show that they significantly outperform existing strategies in both idleness minimization and against an intelligent intruder model, as well as presenting an examination of robustness to communication failure. Our results also indicate important considerations for future strategy design.

AB - The problem of decentralized multi-robot patrol has previously been approached primarily with hand-designed strategies for minimization of "idlenes" over the vertices of a graph-structured environment. Here we present two lightweight neural network-based strategies to tackle this problem, and show that they significantly outperform existing strategies in both idleness minimization and against an intelligent intruder model, as well as presenting an examination of robustness to communication failure. Our results also indicate important considerations for future strategy design.

KW - cs.RO

U2 - 10.1145/3672608.3707752

DO - 10.1145/3672608.3707752

M3 - Conference Contribution (Conference Proceeding)

T3 - Proceedings of the ACM/SIGAPP Symposium on Applied Computing

SP - 823

EP - 831

BT - SAC '25

PB - Association for Computing Machinery

T2 - 40th ACM/SIGAPP Symposium on Applied Computing

Y2 - 31 March 2025 through 4 April 2025

ER -

Lightweight Decentralized Neural Network-Based Strategies for Multi-Robot Patrolling

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Handle.net

Fingerprint

Cite this