Skip to main navigation Skip to search Skip to main content

On partially observable MDPs and BDI models

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Decision theoretic planning in ai by means of solving Partially Observable Markov decision processes (pomdps) has been shown to be both powerful and versatile. However, such approaches are computationally hard and, from a design stance, are not necessarily intuitive for conceptualising many problems. We propose a novel method for solving pomdps, which provides a designer with a more intuitive means of specifying pomdp planning problems. In particular, we investigate the relationship between pomdp planning theory and belief-desire-intention (bdi) agent theory. The idea is to view a bdi agent as a specification of an pomdp problem. This view is to be supported by a correspondence between an pomdp problem and a bdi agent. In this paper, we outline such a correspondence between pomdp and bdi by explaining how to specify one in terms of the other. Additionally, we illustrate the significance of a correspondence by showing empirically that it yields satisfying results in complex domains.

Original languageEnglish
Title of host publicationFoundations and Applications of Multi-Agent Systems - UKMAS Workshops 1996-2000, Selected Papers
PublisherSpringer Verlag
Pages243-259
Number of pages17
ISBN (Print)3540439625, 9783540439622
DOIs
Publication statusPublished - 2002
Externally publishedYes
EventUK Workshops on Multi-Agent Systems, UKMAS 1996-2000 -
Duration: 1 Jan 2002 → …

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2403
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceUK Workshops on Multi-Agent Systems, UKMAS 1996-2000
Period01/01/2002 → …

Fingerprint

Dive into the research topics of 'On partially observable MDPs and BDI models'. Together they form a unique fingerprint.

Cite this