The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space

We consider autonomous partially observable Markov decision processes where the control action influences the observation process only. Considering entropy as the cost incurred by the Markov information state process, the optimal observability problem is posed as a Markov decision scheduling problem...

Full description

Bibliographic Details
Main Authors: Rezaeian, M., Vo, Ba-Ngu, Evans, J.
Format: Journal Article
Published: IEEE 2010
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/24789
_version_ 1848751526935265280
author Rezaeian, M.
Vo, Ba-Ngu
Evans, J.
author_facet Rezaeian, M.
Vo, Ba-Ngu
Evans, J.
author_sort Rezaeian, M.
building Curtin Institutional Repository
collection Online Access
description We consider autonomous partially observable Markov decision processes where the control action influences the observation process only. Considering entropy as the cost incurred by the Markov information state process, the optimal observability problem is posed as a Markov decision scheduling problem that minimizes the infinite horizon cost. This scheduling problem is shown to be equivalent to minimization of an entropy measure, called estimation entropy which is related to the invariant measure of the information state.
first_indexed 2025-11-14T07:54:08Z
format Journal Article
id curtin-20.500.11937-24789
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T07:54:08Z
publishDate 2010
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-247892017-09-13T15:12:59Z The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space Rezaeian, M. Vo, Ba-Ngu Evans, J. observability Estimation entropy sensor scheduling partially observable Markov decision processes We consider autonomous partially observable Markov decision processes where the control action influences the observation process only. Considering entropy as the cost incurred by the Markov information state process, the optimal observability problem is posed as a Markov decision scheduling problem that minimizes the infinite horizon cost. This scheduling problem is shown to be equivalent to minimization of an entropy measure, called estimation entropy which is related to the invariant measure of the information state. 2010 Journal Article http://hdl.handle.net/20.500.11937/24789 10.1109/TAC.2010.2074231 IEEE restricted
spellingShingle observability
Estimation entropy
sensor scheduling
partially observable Markov decision processes
Rezaeian, M.
Vo, Ba-Ngu
Evans, J.
The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space
title The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space
title_full The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space
title_fullStr The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space
title_full_unstemmed The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space
title_short The Optimal Observability of Partially Observable Markov Decision Processes: Discrete State Space
title_sort optimal observability of partially observable markov decision processes: discrete state space
topic observability
Estimation entropy
sensor scheduling
partially observable Markov decision processes
url http://hdl.handle.net/20.500.11937/24789