Utilizing BDI features for transactional agent execution

Mingzhong Wang; K Ramamohanarao; A Unruh

doi:10.1109/IAT.2007.62

Back

Utilizing BDI features for transactional agent execution

Conference paper

Open access

Peer reviewed

Utilizing BDI features for transactional agent execution

Mingzhong Wang, K Ramamohanarao and A Unruh

IEEE Computer Society Technical Committee on Intelligent Informatics (TCII), pp.215-221

Intelligent Agent Technology (IAT) Conference, 2007 (Fremont, United States, 02-Nov-2007–05-Nov-2007)

IEEE Computer Society

2008

DOI: https://doi.org/10.1109/IAT.2007.62

Files and links (2)

pdf

PDF - Published Version622.99 kBDownload View

Published VersionPDF - Published Version Open Access

url

https://doi.org/10.1109/IAT.2007.62View

Published Version

Abstract

concurrency control

data structures

exception handling

planning (artificial intelligence)

software agents

transaction processing

Replanning, which discards all alternatives found in previous execution, and step-by-step backtracking, which iterates back through all previous selection decisions, are two extreme ends of exception handling in agent systems. We describe a new "jump" approach to maintaining a trade-off between them. Choice points, which are maintained in a stack, are introduced to record all plans found by the agent's planning procedures. By iterating on the stack, and reasoning about plan characteristics, an agent can find and launch a suitable plan prior to or in parallel with a compensation process for the failed path, thus achieving its goals efficiently in the presence of exceptions. Our approach combines and utilizes several beneficial features of a BDI agent, such as its data structure and deliberation cycle, together with an open nested transaction system which supports architectural-level concurrency control and distribution management. © 2007 IEEE.

Details

Title: Utilizing BDI features for transactional agent execution
Authors: Mingzhong Wang (Author) - University of Melbourne
K Ramamohanarao (Author) - University of Melbourne
A Unruh (Author) - University of Melbourne
Contributors: Tsau Young Lin (Editor)
Publication details: IEEE Computer Society Technical Committee on Intelligent Informatics (TCII), pp.215-221
Conference details: Intelligent Agent Technology (IAT) Conference, 2007 (Fremont, United States, 02-Nov-2007–05-Nov-2007)
Publisher: IEEE Computer Society
Date published: 2008
DOI: 10.1109/IAT.2007.62
ISBN: 9780769530277
Copyright note: Copyright © 2008 IEEE. Reproduced here in accordance with the publisher's copyright policy. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Organisation Unit: University of the Sunshine Coast, Queensland; USC Business School - Legacy; School of Science, Technology and Engineering
Language: English
Record Identifier: 99449135002621
Output Type: Conference paper

Metrics

61 File views/ downloads

431 Record Views