Modeling Command and Control in Multi-Agent Systems
TEXAS A AND M UNIV COLLEGE STATION DEPT OF COMPUTER SCIENCE
Pagination or Media Count:
Intelligent agents can be quite useful as entities in combat simulations. Recently, there has been a great deal of research on developing enhanced methods for implementing intelligent agents in combat simulations, such as by introducing models of teamwork and collaborative behavior. However, modeling of command-and-control has lagged behind. Much is known about command-and-control in human tactical decision-making TDM teams from studies in cognitive science and organizational psychology. These studies suggest that human decision-makers tend to follow a Naturalistic Decision-Making process, in which situation awareness plays a key role. Hence command-and-control is heavily focused on information-gathering and information-fusion activities, oriented toward reducing uncertainty and identifying the situation, based on which an appropriate response can be applied or adapted from experience or training. In this paper, we provide a brief survey of multi-agent systems architectures, with a focus on combat simulations, and a survey of the cognitive literature on human situation awareness and tactical decision-making. Then we describe a new computational model for command-and-control in multi-agent systems. Primarily, the model focuses on a procedural representation of situation assessment and attempts to capture the decisions regarding information-gathering and information management activities, though we also discuss how to integrate these activities with other on-going aspects of C2 mission, threat-handling, etc. using prioritization. We then discuss an approach to extending this procedure to a team task, which should automatically generate the interactions and information flow necessary to simulate distributed situation awareness.
- Administration and Management
- Command, Control and Communications Systems