Planning in belief space with a labelled uncertainty graph

Daniel Bryce, Subbarao Kambhampati, David E. Smith

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

Planning in belief space with a Labelled Uncertainty Graph, LUG, is an approach that uses a very compact planning graph to guide search in the space of belief states to construct conformant and contingent plans. A conformant plan is a plan that transitions (without sensing) all possible initial states through possibly non-deterministic actions to a goal state. A contingent plan adds the ability to observe state variables and branch execution. The LUG provides heuristics to guide a regression belief space planner, CAltAlt, and a progression belief space planner, PBSP. The key innovation of the LUG is to compactly represent the optimistic projection of several states (those in a belief state) within a single planning graph. Labels on the actions and literals denote the state projections that include them. We show the improvements of using a LUG by comparing it with a multiple planning graphs (one for each possible world) approach within CAltAlt. We also compare the approach to other conformant planners. Lastly, we outline an algorithm that can adapt the approach to stochastic planning.

Original languageEnglish (US)
Title of host publicationAAAI Workshop - Technical Report
Pages1-6
Number of pages6
VolumeWS-04-08
StatePublished - 2004
Event19th National Conference on Artificial Intelligence - San Jose, CA, United States
Duration: Jul 25 2004Jul 26 2004

Other

Other19th National Conference on Artificial Intelligence
CountryUnited States
CitySan Jose, CA
Period7/25/047/26/04

Fingerprint

Planning
Labels
Innovation
Uncertainty

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Bryce, D., Kambhampati, S., & Smith, D. E. (2004). Planning in belief space with a labelled uncertainty graph. In AAAI Workshop - Technical Report (Vol. WS-04-08, pp. 1-6)

Planning in belief space with a labelled uncertainty graph. / Bryce, Daniel; Kambhampati, Subbarao; Smith, David E.

AAAI Workshop - Technical Report. Vol. WS-04-08 2004. p. 1-6.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bryce, D, Kambhampati, S & Smith, DE 2004, Planning in belief space with a labelled uncertainty graph. in AAAI Workshop - Technical Report. vol. WS-04-08, pp. 1-6, 19th National Conference on Artificial Intelligence, San Jose, CA, United States, 7/25/04.
Bryce D, Kambhampati S, Smith DE. Planning in belief space with a labelled uncertainty graph. In AAAI Workshop - Technical Report. Vol. WS-04-08. 2004. p. 1-6
Bryce, Daniel ; Kambhampati, Subbarao ; Smith, David E. / Planning in belief space with a labelled uncertainty graph. AAAI Workshop - Technical Report. Vol. WS-04-08 2004. pp. 1-6
@inproceedings{3055a9affa404558a3dc67f5707738d7,
title = "Planning in belief space with a labelled uncertainty graph",
abstract = "Planning in belief space with a Labelled Uncertainty Graph, LUG, is an approach that uses a very compact planning graph to guide search in the space of belief states to construct conformant and contingent plans. A conformant plan is a plan that transitions (without sensing) all possible initial states through possibly non-deterministic actions to a goal state. A contingent plan adds the ability to observe state variables and branch execution. The LUG provides heuristics to guide a regression belief space planner, CAltAlt, and a progression belief space planner, PBSP. The key innovation of the LUG is to compactly represent the optimistic projection of several states (those in a belief state) within a single planning graph. Labels on the actions and literals denote the state projections that include them. We show the improvements of using a LUG by comparing it with a multiple planning graphs (one for each possible world) approach within CAltAlt. We also compare the approach to other conformant planners. Lastly, we outline an algorithm that can adapt the approach to stochastic planning.",
author = "Daniel Bryce and Subbarao Kambhampati and Smith, {David E.}",
year = "2004",
language = "English (US)",
volume = "WS-04-08",
pages = "1--6",
booktitle = "AAAI Workshop - Technical Report",

}

TY - GEN

T1 - Planning in belief space with a labelled uncertainty graph

AU - Bryce, Daniel

AU - Kambhampati, Subbarao

AU - Smith, David E.

PY - 2004

Y1 - 2004

N2 - Planning in belief space with a Labelled Uncertainty Graph, LUG, is an approach that uses a very compact planning graph to guide search in the space of belief states to construct conformant and contingent plans. A conformant plan is a plan that transitions (without sensing) all possible initial states through possibly non-deterministic actions to a goal state. A contingent plan adds the ability to observe state variables and branch execution. The LUG provides heuristics to guide a regression belief space planner, CAltAlt, and a progression belief space planner, PBSP. The key innovation of the LUG is to compactly represent the optimistic projection of several states (those in a belief state) within a single planning graph. Labels on the actions and literals denote the state projections that include them. We show the improvements of using a LUG by comparing it with a multiple planning graphs (one for each possible world) approach within CAltAlt. We also compare the approach to other conformant planners. Lastly, we outline an algorithm that can adapt the approach to stochastic planning.

AB - Planning in belief space with a Labelled Uncertainty Graph, LUG, is an approach that uses a very compact planning graph to guide search in the space of belief states to construct conformant and contingent plans. A conformant plan is a plan that transitions (without sensing) all possible initial states through possibly non-deterministic actions to a goal state. A contingent plan adds the ability to observe state variables and branch execution. The LUG provides heuristics to guide a regression belief space planner, CAltAlt, and a progression belief space planner, PBSP. The key innovation of the LUG is to compactly represent the optimistic projection of several states (those in a belief state) within a single planning graph. Labels on the actions and literals denote the state projections that include them. We show the improvements of using a LUG by comparing it with a multiple planning graphs (one for each possible world) approach within CAltAlt. We also compare the approach to other conformant planners. Lastly, we outline an algorithm that can adapt the approach to stochastic planning.

UR - http://www.scopus.com/inward/record.url?scp=32144455736&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=32144455736&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:32144455736

VL - WS-04-08

SP - 1

EP - 6

BT - AAAI Workshop - Technical Report

ER -