LiU Electronic Press
Full-text not available in DiVA
Author:
Warnquist, Håkan (Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab) (Linköping University, The Institute of Technology) (APD)
Kvarnström, Jonas (Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab) (Linköping University, The Institute of Technology) (APD)
Doherty, Patrick (Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab) (Linköping University, Department of Computer and Information Science, UASTECH - Autonomous Unmanned Aircraft Systems Technologies) (Linköping University, The Institute of Technology)
Title:
Exploiting Fully Observable and Deterministic Structures in Goal POMDPs
Department:
Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab
Linköping University, The Institute of Technology
Linköping University, Department of Computer and Information Science, UASTECH - Autonomous Unmanned Aircraft Systems Technologies
Publication type:
Conference paper, Presentation (Refereed)
Language:
English
In:
Proceedings of the 23rd International Conference on Automated Planning and Scheduling (ICAPS)
Editor:
Daniel Borrajo, Subbarao Kambhampati, Angelo Oddi, Simone Fratini
Conference:
23rd International Conference on Automated Planning and Scheduling (ICAPS 2013), 10-14 June 2013, Rom, Italy
Publisher: AAAI Press
Year of publ.:
2013
URI:
urn:nbn:se:liu:diva-88637
Permanent link:
http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-88637
ISBN:
978-1-57735-609-7
Subject category:
Computer Science
SVEP category:
Computer science
Abstract(en) :

When parts of the states in a goal POMDP are fully observable and some actions are deterministic it is possibleto take advantage of these properties to efficiently generate approximate solutions. Actions that deterministically affect the fully observable component of the world state can be abstracted away and combined into macro actions, permitting a planner to converge more quickly. This processing can be separated from the main search procedure, allowing us to leverage existing POMDP solvers. Theoretical results show how a POMDP can be analyzed to identify the exploitable properties and formal guarantees are provided showing that the use of macro actions preserves solvability. The efficiency of the method is demonstrated with examples when used in combination with existing POMDP solvers.

Available from:
2013-02-14
Created:
2013-02-14
Last updated:
2013-08-29
Statistics:
57 hits