Attributed Transition-Based Domain Control Knowledge for Domain-Independent Planning (Extended Abstract)

Abstract

This extended abstract from the area of automated planning discusses work on Attributed Transition-Based Domain Control Knowledge (ATB-DCK). ATB-DCK, roughly speaking, represents the "grammar" of solution plans that guides the search. ATB-DCK is expressed by a finite state automaton with attributed states, referring to specific states of objects, connected by transitions imposing constraints on action applicability. This representation stays on side of the planning domain model, but it can be compiled into a classical planning task and thus it complements domain-independent planning techniques. Results on several benchmark domains from the International Planning Competitions show that the use of ATB-DCK often considerably improves efficiency of existing state-of-the-art planning engines.