A Distributed Anytime Algorithm for Dynamic Task Allocation in Multi-Agent Systems
A Distributed Anytime Algorithm for Dynamic Task Allocation in Multi-Agent Systems
We introduce a novel distributed algorithm for multi-agent task allocation problems where the sets of tasks and agents constantly change over time. We build on an existing anytime algorithm (fast-max-sum), and give it significant new capabilities: namely, an online pruning procedure that simplifies the problem, and a branch-and-bound technique that reduces the search space. This allows us to scale to problems with hundreds of tasks and agents. We empirically evaluate our algorithm against established benchmarks and find that, even in such large environments, a solution is found up to 31% faster, and with up to 23% more utility, than state-of-the-art approximation algorithms. In addition, our algorithm sends up to 30% fewer messages than current approaches when the set of agents or tasks changes.
978-1-57735-507-6
701-706
Macarthur, Kathryn
4c7db797-1679-4fd1-9dac-26f84bd5debd
Stranders, Ruben
cca79d07-0668-4231-a80f-5fae6617644c
Ramchurn, Sarvapali
1d62ae2a-a498-444e-912d-a6082d3aaea3
Jennings, Nick
ab3d94cc-247c-4545-9d1e-65873d6cdb30
7 August 2011
Macarthur, Kathryn
4c7db797-1679-4fd1-9dac-26f84bd5debd
Stranders, Ruben
cca79d07-0668-4231-a80f-5fae6617644c
Ramchurn, Sarvapali
1d62ae2a-a498-444e-912d-a6082d3aaea3
Jennings, Nick
ab3d94cc-247c-4545-9d1e-65873d6cdb30
Macarthur, Kathryn, Stranders, Ruben, Ramchurn, Sarvapali and Jennings, Nick
(2011)
A Distributed Anytime Algorithm for Dynamic Task Allocation in Multi-Agent Systems.
Twenty-Fifth Conference on Artificial Intelligence (AAAI), San Francisco, United States.
07 - 11 Aug 2011.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
We introduce a novel distributed algorithm for multi-agent task allocation problems where the sets of tasks and agents constantly change over time. We build on an existing anytime algorithm (fast-max-sum), and give it significant new capabilities: namely, an online pruning procedure that simplifies the problem, and a branch-and-bound technique that reduces the search space. This allows us to scale to problems with hundreds of tasks and agents. We empirically evaluate our algorithm against established benchmarks and find that, even in such large environments, a solution is found up to 31% faster, and with up to 23% more utility, than state-of-the-art approximation algorithms. In addition, our algorithm sends up to 30% fewer messages than current approaches when the set of agents or tasks changes.
Text
macarthur.pdf
- Version of Record
More information
Published date: 7 August 2011
Additional Information:
Event Dates: August 7-11, 2011
Venue - Dates:
Twenty-Fifth Conference on Artificial Intelligence (AAAI), San Francisco, United States, 2011-08-07 - 2011-08-11
Organisations:
Agents, Interactions & Complexity
Identifiers
Local EPrints ID: 272233
URI: http://eprints.soton.ac.uk/id/eprint/272233
ISBN: 978-1-57735-507-6
PURE UUID: 44468b5e-2d3a-49b0-a9d6-d6e72baf398b
Catalogue record
Date deposited: 28 Apr 2011 11:16
Last modified: 15 Mar 2024 03:22
Export record
Contributors
Author:
Kathryn Macarthur
Author:
Ruben Stranders
Author:
Sarvapali Ramchurn
Author:
Nick Jennings
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