Knowledge Representation for Decision Making Agents

Abstract

This technical memorandum documents the implementation of using agent knowledge to inform dynamic behaviors within COMBATXXI. This project partially addresses TRAC research requirements 4.1, 4.5, 4.9, 4.10, 4.13, 4.18, and 4.19 under Support for TRAC M&S and Scenario Enterprise. The results are a methodology that can be used to expand the agility and responsiveness of COMBATXXI for future study use. We demonstrate it is possible to implement non-prescriptive dynamic behavior through use of a knowledge map without modification to the COMBATXXI base code. We also show there are several secondary benefits to this methodology, including representing collective awareness.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 15, 2013
Accession Number
ADA589932

Entities

People

  • David Ohmen
  • Jonathan K. Alt
  • Kyle Quinnell
  • Mario Torres
  • Peter A. Nesbitt
  • Tom Anderson

Organizations

  • United States Army Training and Doctrine Command

Tags

Communities of Interest

  • Air Platforms
  • Autonomy
  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Aircrafts
  • Behavior Control
  • Combat Simulations
  • Database Management Systems
  • Department Of Defense
  • Force Protection
  • Information Operations
  • Instructions
  • Knowledge Management
  • Language
  • Military Operations
  • Normal Distribution
  • Probability
  • Random Walk
  • Scripting Languages
  • Simulations
  • Urban Areas

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computational Modeling and Simulation