AI Planning in a Constraint Programming Framework

摘要

Conventional methods for AI planning use highly specific representations and search methods that can hardly be adapted or extended. Recently, approaches based on more general search frameworks like propositional satisfiability, operations research and constraint programming have been developed. This paper presents a model for domain-independent planning based on an extension of constraint programming. The extension makes it possible to explore the search space without the need to focus on plan length, and to favor other criteria like resource-related properties.