Artificial Intelligence: Expert Systems for Corps Tactical Planning and Other Applications

摘要

Systems such as those that play games, diagnose engine problems, or organize cargo loads in ships are all examples of artificial intelligence. The sub-discipline of expert systems deals with computerized imitation of the reasoning of judgment process of human experts. CECOM's expert system for tactical planning draws its tactical expertise from the US Army War College students that comprise study group. Computer scientists who have a long-term commitment to the CECOM project work with the study group experts to extract and understand what rules, guidelines, or thought processes the group uses to generate a tactical plan for a Corps operation. The computer scientist takes these lists of information and converts them into computer knowledge which eventually becomes rules that will govern program output. In the heuristic environment of tactical planning, it would not be unreasonable to expect the finished expert system to contain between 20 and 50,000 such rules. The multi-year approach to the project is driven home by the recognition that a hand-crafted expert system might have 20 rules after the first year of effort. By working with CECOM's knowledge engineers, participating in numerous group sessions, going on TDY trips, and reading extensively, study group members developed an appreciation for some of the difficulties and opportunities associated with the use of artificial intelligence in its various military applications.