Ant-inspired dynamic task allocation via gossiping

Hsin Hao Su, Lili Su, Anna Dornhaus, Nancy Lynch

Research output: ResearchConference contribution

Abstract

We study the distributed task allocation problem in multi-agent systems, where each agent selects a task in such a way that, collectively, they achieve a proper global task allocation. In this paper, inspired by specialization on division of labor in ant colonies, we propose several scalable and efficient algorithms to dynamically allocate the agents as the task demands change. The algorithms have their own pros and cons, with respect to (1) how fast they react to dynamic demands change, (2) how many agents need to switch tasks, (3) whether extra agents are needed, and (4) whether they are resilient to faults.

LanguageEnglish (US)
Title of host publicationStabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings
PublisherSpringer Verlag
Pages157-171
Number of pages15
Volume10616 LNCS
ISBN (Print)9783319690834
DOIs
StatePublished - 2017
Event19th International Symposium on Stabilization, Safety, and Security of Distributed Systems, SSS 2017 - Boston, United States
Duration: Nov 5 2017Nov 8 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10616 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other19th International Symposium on Stabilization, Safety, and Security of Distributed Systems, SSS 2017
CountryUnited States
CityBoston
Period11/5/1711/8/17

Fingerprint

Gossiping
Task Allocation
Specialization
Ant Colony
Multi-agent Systems
Division
Switch
Fault
Efficient Algorithms
Multi agent systems
Switches
Personnel

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Su, H. H., Su, L., Dornhaus, A., & Lynch, N. (2017). Ant-inspired dynamic task allocation via gossiping. In Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings (Vol. 10616 LNCS, pp. 157-171). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10616 LNCS). Springer Verlag. DOI: 10.1007/978-3-319-69084-1_11

Ant-inspired dynamic task allocation via gossiping. / Su, Hsin Hao; Su, Lili; Dornhaus, Anna; Lynch, Nancy.

Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings. Vol. 10616 LNCS Springer Verlag, 2017. p. 157-171 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10616 LNCS).

Research output: ResearchConference contribution

Su, HH, Su, L, Dornhaus, A & Lynch, N 2017, Ant-inspired dynamic task allocation via gossiping. in Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings. vol. 10616 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10616 LNCS, Springer Verlag, pp. 157-171, 19th International Symposium on Stabilization, Safety, and Security of Distributed Systems, SSS 2017, Boston, United States, 11/5/17. DOI: 10.1007/978-3-319-69084-1_11
Su HH, Su L, Dornhaus A, Lynch N. Ant-inspired dynamic task allocation via gossiping. In Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings. Vol. 10616 LNCS. Springer Verlag. 2017. p. 157-171. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Available from, DOI: 10.1007/978-3-319-69084-1_11
Su, Hsin Hao ; Su, Lili ; Dornhaus, Anna ; Lynch, Nancy. / Ant-inspired dynamic task allocation via gossiping. Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings. Vol. 10616 LNCS Springer Verlag, 2017. pp. 157-171 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inbook{24c4c7af4e9a4ca5a97ce858210258bf,
title = "Ant-inspired dynamic task allocation via gossiping",
abstract = "We study the distributed task allocation problem in multi-agent systems, where each agent selects a task in such a way that, collectively, they achieve a proper global task allocation. In this paper, inspired by specialization on division of labor in ant colonies, we propose several scalable and efficient algorithms to dynamically allocate the agents as the task demands change. The algorithms have their own pros and cons, with respect to (1) how fast they react to dynamic demands change, (2) how many agents need to switch tasks, (3) whether extra agents are needed, and (4) whether they are resilient to faults.",
author = "Su, {Hsin Hao} and Lili Su and Anna Dornhaus and Nancy Lynch",
year = "2017",
doi = "10.1007/978-3-319-69084-1_11",
isbn = "9783319690834",
volume = "10616 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "157--171",
booktitle = "Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings",
address = "Germany",

}

TY - CHAP

T1 - Ant-inspired dynamic task allocation via gossiping

AU - Su,Hsin Hao

AU - Su,Lili

AU - Dornhaus,Anna

AU - Lynch,Nancy

PY - 2017

Y1 - 2017

N2 - We study the distributed task allocation problem in multi-agent systems, where each agent selects a task in such a way that, collectively, they achieve a proper global task allocation. In this paper, inspired by specialization on division of labor in ant colonies, we propose several scalable and efficient algorithms to dynamically allocate the agents as the task demands change. The algorithms have their own pros and cons, with respect to (1) how fast they react to dynamic demands change, (2) how many agents need to switch tasks, (3) whether extra agents are needed, and (4) whether they are resilient to faults.

AB - We study the distributed task allocation problem in multi-agent systems, where each agent selects a task in such a way that, collectively, they achieve a proper global task allocation. In this paper, inspired by specialization on division of labor in ant colonies, we propose several scalable and efficient algorithms to dynamically allocate the agents as the task demands change. The algorithms have their own pros and cons, with respect to (1) how fast they react to dynamic demands change, (2) how many agents need to switch tasks, (3) whether extra agents are needed, and (4) whether they are resilient to faults.

UR - http://www.scopus.com/inward/record.url?scp=85032700544&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032700544&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-69084-1_11

DO - 10.1007/978-3-319-69084-1_11

M3 - Conference contribution

SN - 9783319690834

VL - 10616 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 157

EP - 171

BT - Stabilization, Safety, and Security of Distributed Systems - 19th International Symposium, SSS 2017, Proceedings

PB - Springer Verlag

ER -