High-Level Symbolic Representation for Intelligent Agents Across Multiple Architectures

摘要

This report describes the High-bevel Symbolic Representation (HLSR) project for the U.S. Air Force PRDA 03-01-HE: Human Performance in Modeling and Simulation, Technical Area 2: Opposing Force Behaviors. This report summarizes the work done on Defense Modeling Simulation contract F33615-03-C-6343 to develop a high level symbolic representation (HLSR) for behavior modeling. This effort seeks to increase development efficiency and reuse in behavior modeling. The report describes the development of a high level language that abstracts the details of individual intelligent system architectures (ISA), allowing developers to focus their effort on tasks directly related to producing intelligent behavior. This language is designed to be complied into executable representations on multiple ISAs. This report targets two ISAs, Soar and ACT-R. These ISAs have a proven tract record of generating capable behavior models in many domains. There were three primary goals. First, the desire to construct a specification for HLSR sufficient for its use in behavior modeling and for compiler design and implementation. Second, the desire to define mappings and probable transformation processes from HLSR to Soar and ACT-R. Third, the desire to prove the feasibility of this approach by demonstrating how HLSR solved real problems faced by knowledge engineers and how an HLSR model, which solved these problems, could be compiled to an ISA.