Real Tme Intelligent Coaching for Command and Control

Period of Performance: 09/01/1998 - 06/01/1999

$99.9K

Phase 1 STTR

Recipient Firm

Aptima, Inc.
12 Gill Street Array
Woburn, MA 01801
Principal Investigator
Firm POC

Research Institution

University of Georgia
Office for Sponsored Programs
Athens, GA 30602
Institution POC

Research Topics

Abstract

Technology is changing the face of warfare, placing ever increasing demands on the speed, scope, and accuracy of command andTechnology is changing the face of warfare, placing ever increasing demands on the speed, scope, and accuracy of command and control decisions. We propose to develop and test intelligent coaching strategies based on context-sensitive models of C2 decision making, and to test them in a dynamic C2 team simulation. Our approach blends formal decision models with intuitive explanations of model results and recommendations, using natural language that is meaningful to both experts and non-experts within a domain as well a graphical representations. We will focus on the decisions made by individuals acting as part of a command and control team. Our proposed test environment for intelligent coaching concepts is a well-developed and tested multi-user simulation package--the Distributed Dynamic Decisionmaking (DDD) C2 team simulation. At the end of Phase I, we will demonstrate an agent-based coach integrated into the DDD environment and provide model-driven quantitative measures of performance. In Phase II, we will build on the results of Phase I to conduct quantitative tests of decision aid effectiveness in the DDD environment, and implement the decision aid concepts that have proved to be most successful in the DDD for demonstration and quantitative testing in higher fidelity simulations or real-world environments.