Distributed Global Air and Missile Defense Resource Management

The management of air and missile defense resources has long been understood to be one of the critical technologies for missile defense. Partly because of its complexity, the problem has in the past been broken into subproblems to be solved separately and then pieced together. For a variety of reasons, it is now both possible and necessary to treat missile defense resource management with an integrated approach. Among the reasons are the potential for more effective and economical use of resources, the emergence of a system of systems for missile defense, the growth in computing power, and the development of new algorithmic approaches for dealing with the complexity of the problem. We propose a new algorithmic approach to achieve integrated distributed global missile defense resource management. Each agent is responsible for and controls a collection of sensors and shooters, and has detailed knowledge of their capabilities, but has limited knowledge of the capabilities of resources belonging to other agents. Each agent uses simulation-based approximate dynamic programming to calculate the values of possible actions in the current state and fuses them with values calculated by other agents. Agents coordinate actions by a token passing scheme.