Solving Large-Scale Planning Problems by Decomposition and Macro Generation

Abstract

Large-scale classical planning problems such as factory as- sembly problems pose a significant challenge for domain- independent planners. We propose a macro-based planner which automatically identifies subproblems, generate macros from subplans and integrate the subplans by solving the whole augmented problem. We show experimentally that our approach can be used to solve large problem instances that are beyond the reach of current state-of-the-art satisficing plan- ners.