رویکرد منطق فازی برای تجزیه و تحلیل اطلاعاتی از انجام ماموریت ‌های شناسایی نظامی واقعی و شبیه سازی شده A Fuzzy Logic Approach for Intelligence Analysis of Actual and Simulated Military Reconnaissance Missions

With modern advances in the computing industry, the Army has committed to fielding an automated force by the turn of the century. General Gordon Sullivan, former Army Chief of Staff, stated his views of the future Army as: “In the] Army of the 21St century, rapad technologacal developments zn anformataon management and process are usherang an what many helaeve to be the hegznnzng of a post-zndustrzal age; the Informataon Age. The macroprocessor as revolutaonazang the way that we lave our laves as zndzvzduals, the way that socaety functaons, and the way that we are lzkely to fight our future wars … . These powerful developments are leadang socaety toward an uncertaan but znterestzng future; a future whach at as just begznnang to explore [I] ’’ In accordance with this vision, the Army of the 21St century will be organized into combat units which are driven by information and information technologies. These networked units will possess the capability to rapidly and effectively react to the volatility associated with the modern battlefield. The Army’s tactical operations center (TOC) will evolve into an information warehouse consisting of numerous command and control (C2) systems which will provide decision-makers with access to a high-fidelity conceptual view of the battlefield. These new systems will undoubtedly create new challenges for future commanders and their staffs. Specifically, systems which merely increase the amount of data the staffs must process are likely to have deleterious effects on their performance. With the expected increased information flow, the challenge becomes the staffs’ ability to decipher the information that is critical, and to effectively apply this information in the decision process. Unfortunately, unless future systems can help analyze and synthesize this data, the tendency will be for staffs to become overwhelmed with small details, thus losing sight of the “big picture.” To meet the requirements of the Army of the 21St century, future C2 systems must go beyond the capabilities of today’s information systems. These systems must incorporate artificial intelligence (AI)-based decision support tools so that they become a part of the staff, rather than just a tool used by the staff. This will enable commanders and their staffs to generate, modify, and analyze complete, consistent, and robust plans in real-world, resource-constrained environments. This paper describes a technology – fuzzy logic – that could have a significant impact on the Army’s many automation initiatives. Fuzzy logic was introduced over 20 years ago as an approach to generalize classical two-value logic for reasoning under uncertainty [2]. The environment in which military operations are planned and executed is fraught with uncertainty and imprecision. Noted military theorist Carl von Clauswitz once stated that “many intelligence reports in war are contradictory; even more are false, and most are uncertain [3].” Even though these words were written over 160 years ago, they still ring true today. Based on the statements above, technolo gies that handle uncertainty and imprecision should play an important role in the development of future C2 systems. Fortunately, systems which are based on fuzzy logic technologies are inherently able to tolerate uncertainty [4]. Specifically, the uncertain and imprecise knowledge that military planners use can be effectively represented using fuzzy linguistic variables. These variables use linguistic labels which are readily understood by the domain experts and are useful for communicating concepts and knowledge with human beings [5]. Additionally, they have a well-defined quantitative component that allows for interoperability among other non-fuzzy Army systems. Representing knowledge in this manner simplifies the knowledge acquisition process, thereby providing a more consistent result among multiple users. Conventional rule-based or statistical approaches are other possible solutions to this problem. However, a fuzzy logic approach provides an alternative technology that alleviates many of the difficulties conventional approaches have in handling uncertainty, imprecision, and ambiguity. Secondly, by using fuzzy logic techniques the system can readily represent and reason with both explicit information and the various types of uncertainty that are implicit in much of this knowledge. Finally, fuzzy logic methodologies provide a “communication bridge” between domain experts and the system through fuzzy linguistic variables.