Representing Multi-Agent Planning in CLP

Abstract

We explore the use of Constraint Logic Programming (CLP) as a platform for experimenting with planning domains in presence of multiple interacting agents. We develop a novel constraint-based action language, ℬ MAPthat enables the declarative description of large classes of multi-agent and multi-valued domains. ℬ MAP supports several complex features, including combined effects, concurrency constraint, interacting actions, and delayed effects.