# Angelic Hierarchical Planning: Optimal and Online Algorithms (Revised)

### Bhaskara Marthi, Stuart J. Russell and Jason Wolfe

###
EECS Department

University of California, Berkeley

Technical Report No. UCB/EECS-2009-122

August 22, 2009

### http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-122.pdf

High-level actions (HLAs) are essential tools for coping with the large search spaces and long decision horizons encountered in real-world decision making. In a recent paper, we proposed an "angelic" semantics for HLAs that supports proofs that a high-level plan will (or will not) achieve a goal, without first reducing the plan to primitive action sequences. This paper extends the angelic semantics with cost information to support proofs that a high-level plan is (or is not) optimal. We describe the Angelic Hierarchical A* algorithm, which generates provably optimal plans, and show its advantages over alternative algorithms. We also present the Angelic Hierarchical Learning Real-Time A* algorithm for situated agents, one of the first algorithms to do hierarchical lookahead in an online setting. Since high-level plans are much shorter, this algorithm can look much farther ahead than previous algorithms (and thus choose much better actions) for a given amount of computational effort. This is a revised, extended version of a paper by the same name appearing in ICAPS '08.

BibTeX citation:

@techreport{Marthi:EECS-2009-122, Author = {Marthi, Bhaskara and Russell, Stuart J. and Wolfe, Jason}, Title = {Angelic Hierarchical Planning: Optimal and Online Algorithms (Revised)}, Institution = {EECS Department, University of California, Berkeley}, Year = {2009}, Month = {Aug}, URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-122.html}, Number = {UCB/EECS-2009-122}, Abstract = {High-level actions (HLAs) are essential tools for coping with the large search spaces and long decision horizons encountered in real-world decision making. In a recent paper, we proposed an "angelic" semantics for HLAs that supports proofs that a high-level plan will (or will not) achieve a goal, without first reducing the plan to primitive action sequences. This paper extends the angelic semantics with cost information to support proofs that a high-level plan is (or is not) optimal. We describe the Angelic Hierarchical A* algorithm, which generates provably optimal plans, and show its advantages over alternative algorithms. We also present the Angelic Hierarchical Learning Real-Time A* algorithm for situated agents, one of the first algorithms to do hierarchical lookahead in an online setting. Since high-level plans are much shorter, this algorithm can look much farther ahead than previous algorithms (and thus choose much better actions) for a given amount of computational effort. This is a revised, extended version of a paper by the same name appearing in ICAPS '08.} }

EndNote citation:

%0 Report %A Marthi, Bhaskara %A Russell, Stuart J. %A Wolfe, Jason %T Angelic Hierarchical Planning: Optimal and Online Algorithms (Revised) %I EECS Department, University of California, Berkeley %D 2009 %8 August 22 %@ UCB/EECS-2009-122 %U http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-122.html %F Marthi:EECS-2009-122