The use of abstraction in problem solving is an eeective approach to reducing search, but nding good abstractions is a diicult problem. The rst algorithm that completely automates the generation of abstraction hierarchies is Knoblock's ALPINE, but this algorithm is only able to automatically abstract the preconditions of operators. In this paper we present an algorithm that automatically abstracts not only the preconditions but also the eeects of operators, and produces ner-grained abstraction hierarchies than ALPINE. The same algorithm also formalizes and selects the primary eeects of operators , which is thus useful even for planning without abstraction. We present a theorem that describes the necessary and suu-cient conditions for a planner to be complete, when guided by primary eeects.
[1]
Earl D. Sacerdoti,et al.
Planning in a Hierarchy of Abstraction Spaces
,
1974,
IJCAI.
[2]
Austin Tate,et al.
Generating Project Networks
,
1977,
IJCAI.
[3]
David E. Wilkins,et al.
Domain-Independent Planning: Representation and Plan Generation
,
1984,
Artif. Intell..
[4]
Qiang Yang,et al.
ABTWEAK: Abstracting a Nonlinear, Least Commitment Planner
,
1990,
AAAI.
[5]
Qiang Yang,et al.
Characterizing Abstraction Hierarchies for Planning
,
1991,
AAAI.
[6]
Craig A. Knoblock.
Automatically generating abstractions for problem solving
,
1991
.