Share Email Print

Proceedings Paper

Knowledge focus via software agents
Author(s): Donald E. Henager
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The essence of military Command and Control (C2) is making knowledge intensive decisions in a limited amount of time using uncertain, incorrect, or outdated information. It is essential to provide tools to decision-makers that provide: * Management of friendly forces by treating the "friendly resources as a system". * Rapid assessment of effects of military actions againt the "enemy as a system". * Assessment of how an enemy should, can, and could react to friendly military activities. Software agents in the form of mission agents, target agents, maintenance agents, and logistics agents can meet this information challenge. The role of each agent is to know all the details about its assigned mission, target, maintenance, or logistics entity. The Mission Agent would fight for mission resources based on the mission priority and analyze the effect that a proposed mission's results would have on the enemy. The Target Agent (TA) communicates with other targets to determine its role in the system of targets. A system of TAs would be able to inform a planner or analyst of the status of a system of targets, the effect of that status, adn the effect of attacks on that system. The system of TAs would also be able to analyze possible enemy reactions to attack by determining ways to minimize the effect of attack, such as rerouting traffic or using deception. The Maintenance Agent would scheudle maintenance events and notify the maintenance unit. The Logistics Agent would manage shipment and delivery of supplies to maintain appropriate levels of weapons, fuel and spare parts. The central idea underlying this case of software agents is knowledge focus. Software agents are createad automatically to focus their attention on individual real-world entities (e.g., missions, targets) and view the world from that entities perspective. The agent autonomously monitors the entity, identifies problems/opportunities, formulates solutions, and informs the decision-maker. The agent must be able to communicate to receive and disseminate information and provide the decision-maker with assistance via focused knowledge. THe agent must also be able to monitor the state of its own environment and make decisions necessary to carry out its delegated tasks. Agents bring three elements to the C2 domain that offer to improve decision-making. First, they provide higher-quality feedback and provide it more often. In doing so, the feedback loop becomes nearly continuous, reducing or eliminating delays in situation updates to decision-makers. Working with the most current information possible improves the control process, thus enabling effects based operations. Second, the agents accept delegation of actions and perform those actions following an established process. Agents' consistent actions reduce the variability of human input and stabilize the control process. Third, through the delegation of actions, agents ensure 100 percent consideration of plan details.

Paper Details

Date Published: 19 September 2001
PDF: 8 pages
Proc. SPIE 4367, Enabling Technology for Simulation Science V, (19 September 2001); doi: 10.1117/12.440046
Show Author Affiliations
Donald E. Henager, Science Applications International Corp. (United States)

Published in SPIE Proceedings Vol. 4367:
Enabling Technology for Simulation Science V
Alex F. Sisti; Dawn A. Trevisani, Editor(s)

© SPIE. Terms of Use
Back to Top