RAO*: an Algorithm for Chance-Constrained POMDP's

• Main Authors: Santana, Pedro (Contributor), Thiebaux, Sylvie (Author), Williams, Brian Charles (Contributor)
• Other Authors: Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory (Contributor), Massachusetts Institute of Technology. Department of Aeronautics and Astronautics (Contributor)
• Format: Article
• Language: English
• Published: Association for the Advancement of Artificial Intelligence, 2016-03-02T23:12:33Z.
• Subjects: Article
• Online Access: Get fulltext

Autonomous agents operating in partially observable stochastic environments often face the problem of optimizing expected performance while bounding the risk of violating safety constraints. Such problems can be modeled as chance-constrained POMDP's (CC-POMDP's). Our first contribution is a systematic derivation of execution risk in POMDP domains, which improves upon how chance constraints are handled in the constrained POMDP literature. Second, we present RAO*, a heuristic forward search algorithm producing optimal, deterministic, finite-horizon policies for CC-POMDP's. In addition to the utility heuristic, RAO* leverages an admissible execution risk heuristic to quickly detect and prune overly-risky policy branches. Third, we demonstrate the usefulness of RAO* in two challenging domains of practical interest: power supply restoration and autonomous science agents
United States. Air Force Office of Scientific Research (Grant FA95501210348)
United States. Air Force Office of Scientific Research (Grant FA2386-15-1-4015)
SUTD-MIT Graduate Fellows Program
NICTA