Combined Sensor/Interceptor Scheduling based on AI Techniques

The ultimate goal of this proposed effort is to improve the BMDS sensor and interceptor real-time scheduling capability. These improvements will include more efficient use of sensor and interceptor resources, increased kill probabilities, reduced probability of leakage, improved tracking accuracy, the ability to efficiently handle a large number of targets and resources, and faster scheduling and rescheduling. This effort builds on our previous successful interceptor scheduling work for the MDA and scheduling the Air Force"s Space Surveillance Network (SSN). Those efforts demonstrated separately our ability to more optimally schedule ballistic missile intercept (using fewer interceptors to achieve equal or reduced Total Residual Leakage (TRL)) than competing algorithms and to achieve far smaller tracking error covariances with the same set of sensors than the current SSN sensor scheduling algorithms. This effort will combine the techniques from those projects and others to create a single system that will simultaneously optimize the tasking of sensors and interceptors for a solution more optimal than would be reached by optimizing them separately. The feasibility of our approach will be proven by developing a prototype in Phase I. Lockheed Martin, lead system integrator for both C2BMC and AEGIS BMD, strongly supports this effort for transition.