Fast Downward Making use of causal dependencies in the problem representation

Fast Downward is a propositional planning system based on heuristic search. Compared to other heuristic planners such as FF or HSP, it has two distinguishing features: First, it is tailored towards planning tasks with non-binary (but nite domain) state variables. Second, it exploits the causal de- pendency between state variables to solve relaxed planning problems in a hierarchical fashion. Fast Downward is a planning system based on heuristic state space search, in the spirit of HSP or FF (Bonet & Geffner 2001; Hoffmann & Nebel 2001). It makes use of the causal graph (or CG) heuristic, introduced in an ICAPS 2004 paper (Helmert 2004). In this extended abstract, we aim at providing a high-level overview of Fast Downward, emphasizing the features that are not described in the CG article. While the CG heuristic was introduced for pure STRIPS domains, Fast Downward is capable of dealing with the complete propositional, non-temporal part of PDDL. In other words, it handles arbitrary ADL constructs and derived predicates (axioms).