Action localness, genericity and invariants in STRIPS

NY Foo*, A Nayak, M Pagnucco, P Peppas, Yan Zhang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

The STRIPS specification for actions, despite criticisms of its shortcomings, remains popular in implementations because of its simplicity and intuitive appeal. Lifschitz showed in his logical account of STRIPS that states can safely he represented as sets of ground atom, and this is our starting point. Our aim is to provide logical foundations for design heuristics and procedural implementations of STRIPS. In particular, we formalize the notions of 'localness' in pre-and post-conditions of actions. This justifies and explicates the intuitions that underlie the diagrammatic reasoning typically used to design such conditions. A formal account is also given for 'generic' actions that have parameters. We conclude with a method for extracting candidate action invariants that distinguishes between two kinds of inertia implied by the action specifications.

Original languageEnglish
Title of host publicationIJCAI-95 - proceedings of the 14th international joint conference on artificial intelligence, vols 1 and 2
Place of PublicationSan Francisco, CA
PublisherMORGAN KAUFMANN PUB INC
Pages549-554
Number of pages6
ISBN (Print)1558603638
Publication statusPublished - 1995
Event14th international joint conference on Artificial intelligence: IJCAI'95 - Montreal, Quebec, Canada
Duration: 20 Aug 199525 Aug 1995

Publication series

NameINTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE
PublisherMORGAN KAUFMANN PUB INC
ISSN (Print)1045-0823

Conference

Conference14th international joint conference on Artificial intelligence
CountryCanada
CityMontreal, Quebec
Period20/08/9525/08/95

Cite this

Foo, NY., Nayak, A., Pagnucco, M., Peppas, P., & Zhang, Y. (1995). Action localness, genericity and invariants in STRIPS. In IJCAI-95 - proceedings of the 14th international joint conference on artificial intelligence, vols 1 and 2 (pp. 549-554). (INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE). San Francisco, CA: MORGAN KAUFMANN PUB INC.