Autonomous Agents Design for Digital Network Maximization in Joint C4I Systems

摘要

The advent of the computer age has brought about an ever increasing demand to transfer exponentially increasing amounts of information, and the associated problems of information sharing. The focus of this paper is to best utilize available digital communications assets in a constrained radio frequency (RF) spectrum to allow sufficient transfer of information providing Department of Defense (DOD) assets flexible, rapid, and in-flight reprogramming, re-planning of strike and cruise missile assets, to engage a high value, emergent target, in the shortest possible time. The postulated methods of utilizing autonomous agent type methods to manage information flow across network nodes has applicability to all digital networks. Inspired by the pioneering work of Pattie Maes, at Massachusetts Institute of Technology (MIT), and previous examination of communications node management, the implementation of independent processes, working on behalf of a host system, to optimize the effective meaningful throughput on a communications channel is not only desirable, but necessary. The evolution of semi intelligent software, whether called Artificial Intelligence, Intelligent Agents, or Autonomous Agents, has reached a level of sophistication allowing the insertion of meaningful articulated processes within existing, and future systems to maximize the network efficiency systematically. Recent work by Michael Cohen on Sodabots, and the evolution of user interactive TinyMUDS of the Maas-Neotek family, a virtual type personality environment, has demonstrated the ability of software to deal with dynamic and changing conditions. The exponential increase in micro-processor power has, for the first time, made available the hardware for such agent implementations as compact, self contained, embedded systems, in direct support of larger existing systems.