The use of artificial intelligence in military simulations

We examine the potential of AI techniques to achieve the realism that we desire, and survey the research that members of the Office of Artificial Intelligence Analysis and Evaluation (OAIAE) and Department of Electrical Engineering and Computer Science (D/EECS) at the US Military Academy have conducted in support of this goal. We describe how a hybrid expert system has performed terrain analysis resulting in the development of mission plans that can be integrated as parts of an automated reasoning system for computer generated forces. The entity behaviors resulting from this system are realistic enough to provide acceptable training value for human opponents in a distributed interactive simulation. We examine the problem of 3D representation of high-resolution terrain in simulations and offer suggestions for tessellating polygons to maintain a high degree of fidelity in militarily significant terrain while reducing the number used to represent insignificant terrain. We investigate current developments in automated target recognition. Finally, we examine the application of knowledge-based systems to the testing of material development systems in simulation