ARMS: an automatic knowledge engineering tool for learning action models for AI planning

• DOI: 10.1017/s0269888907001087
• Publication Date: 2007-07-05T11:25:24Z
• Authors: KANGHENG WU; QIANG YANG; YUNFEI JIANG
• Keywords: Solver; Satisfiability

Abstract We present an action model learning system known as ARMS (Action-Relation Modelling System) for automatically discovering models from a set of successfully observed plans. Current artificial intelligence (AI) planners show impressive performance in many real world and domains, but they all require the definition model. is aimed at example plans, where each plan sequence traces. These can then be used by human editors to refine. The expectation that this will lessen burden designing scratch. In paper, we describe detail. To learn models, gathers knowledge on statistical distribution frequent sets actions It builds weighted propositional satisfiability (weighted SAT) problem solves it using MAXSAT solver. Furthermore, empirical evidence indeed good approximation finally effectively.