Abstract
A contingent plan can be encoded as a rooted graph where branching occurs due to sensing. In many applications it is desirable to limit this branching; either to reduce the complexity of the plan (e.g. for subsequent execution by a human), or because sensing itself is deemed to be too expensive. This leads to an established planning problem that we refer to as branching-bounded contingent planning. In this paper, we formalise solutions to such problems in the context of history-, and belief-based policies: under noisy sensing, these policies exhibit differing notions of sensor actions. We also propose a new algorithm, called BAO*, that is able to find optimal solutions via belief space search. This work subsumes both conformant and contingent planning frameworks, and represents the first practical treatment of branching-bounded contingent planning that is valid under partial observability.
Original language | English |
---|---|
Title of host publication | 2nd ICAPS Workshop on Explainable AI Planning (XAIP'19) |
Publisher | Kings College London Planning (KCL Planning) |
Publication status | Published - 2019 |
Event | International Workshop on Explainable AI Planning - Berkeley, United States Duration: 12 Jul 2019 → 12 Jul 2019 Conference number: 2 https://kcl-planning.github.io/XAIP-Workshops/ICAPS_2019 |
Workshop
Workshop | International Workshop on Explainable AI Planning |
---|---|
Abbreviated title | XAIP 2019 |
Country/Territory | United States |
City | Berkeley |
Period | 12/07/19 → 12/07/19 |
Internet address |