Exploiting Fully Observable and Deterministic Structures in Goal POMDPs

Authors

  • Håkan Warnquist Scania and Linköping University
  • Jonas Kvarnström Linköping University
  • Patrick Doherty Linköping University

DOI:

https://doi.org/10.1609/icaps.v23i1.13554

Keywords:

POMDPs, Planning Algorithms, Sequential Decision Making

Abstract

When parts of the states in a goal POMDP are fully observable and some actions are deterministic it is possible to 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.

Downloads

Published

2013-06-02

How to Cite

Warnquist, H., Kvarnström, J., & Doherty, P. (2013). Exploiting Fully Observable and Deterministic Structures in Goal POMDPs. Proceedings of the International Conference on Automated Planning and Scheduling, 23(1), 242-250. https://doi.org/10.1609/icaps.v23i1.13554

Issue

Vol. 23 (2013): Twenty-Third International Conference on Automated Planning and Scheduling

Section

Full Technical Papers