Tactical Decision Making in Intelligent Agents: Developing Autonomous Forces in NPSNET

This thesis presents a conceptual framework, system architecture, and working prototype for a tactical decision making model. This model was developed within the context of intelligent autonomous forces in combat modeling systems. The goal of this model is to realistically portray the behavior of tactical units operating on the battlefield. In our prototype, tactical decision making principles and heuristics are modeled as rules in a logic programming system, and are implemented in an expert system development environment. The current implementation plans, executes, and monitors its decisions in real-time during the course of the combat simulation. This research also examines several challenges in the modeling and execution of tactical-level decisions of an autonomous force. This thesis is a significant first step in developing fully automated forces that model human tactical decision making.