Many Agents Are Better Than One
by Georgeff, Michael P.
Technical Note 417
Institution: AI Center, SRI International
Address: 333 Ravenswood Ave., Menlo Park, CA 94025
Mar 1987.
This paper aims to show how much of the frame problem can be alleviated by using domain models that allow for the simultaneous occurrence of actions and events. First, a generalized situation calculus is constructed for describing and reasoning about events in multiagent settings. Notions of independence and causality are then introduced and it is shown how they can be used to determine the persistence of facts over time. Finally, it is shown how these notions, together with traditional predicate circumscription, make it possible to retain a simple model of action while avoiding most of the difficulties associated with the frame problem.
 Adobe PDF |
 BibTeX |
 EndNote |
