Aircraft & Cruise Missile Mission and Route Planning Using Real-Time Pattern Matching Techniques

In this proposal we present a novel approach to aircraft and missile mission and route planning using neural networks. The approach is unique in four aspects: l) it is a proactive planner that looks at the global topographic characteristics of the battlefield environment when generating plans; 2) it uses a cost surface to determine the advantages or disadvantages of moving through an area by integrating the effects of terrain, weather, and threat into a concise data structure; 3) it uses multi-resolution grid searching to reduce the time required to produce a route, making the best possible use of its computational resources; 4) it is designed specifically for migration into custom VLSI neural network hardware developed at IRS.