Intelligent Agents for Computer-Generated Forces
Abstract
Computer-generated forces (CGF) are an important part of today's training devices. When linked to manned simulators, these computer-generated entities provide a dynamic and realistic environment for interaction of human participants. It also allows the addition of many players, which might not be cost effective using many networked, manned devices as an alternative. These CGFs are comprised of two specific objects: equipment model and a behavioral or cognitive model. The equipment model represents the machine which, in this case, is an aircraft with its associated dynamics, weapons systems, controls and avionics systems. The cognitive model corresponds to how the machine operator, a pilot in this case, reacts in the dynamic environment. This will be based on mission knowledge, tactical doctrine, and situation awareness. Modeling of the cognitive portion of the computer-generated forces has been accomplished using several techniques including classical artificial intelligence(AI) techniques such as SOAR ("talking a State, applying an operator, And generating Results"), other AI formulations such as FuzzyCLIPS and Modular Knowledge, Acquisition Tool (M-KAT), adoption of analytical military models such as Suppressor, and specialized CGFs such as Modular Semi-Automated Force(ModSAF) and Interactive Tactical Environment Management System(ITEMS). This paper overviews these cognitive modeling techniques focusing on 14 specific features associated with intelligent agents.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1996
- Accession Number
- ADA459709
Entities
People
- Ellen Mallery
- Gary R. George
- Marie Pope