SBIR-STTR Award

Technology is a controlling variable for cost, schedule, and performance design of U.S. ballistic missile defense systems. Target discrimination (the ability to identify or engage any one target when multiple targets are present) during National Missile Defense (NMD) midcourse engagement is a complex technological hurdle. Exoatmospheric kill vehicle (EKV) sensors need to discriminate among warhead(s), decoys and penetration aids in an extremely short detect-to-kill time. Feature differences among decoys, penetration aids, and warheads are not adequate for discrimination by current EKV passive IR sensors and the BMC4I surveillance system. Our Fusion-Generated Discrimination (FGD) approach will generate reliable, accurate Target Object Maps (TOM) to overcome future system hit-to-kill limitations. Using multispectral and active/passive sensor inputs, our FGD Toolkit will combine scene and kinematic processes. The FGD subsystem leverages proven multiple-hypothesis fusion and extended Kalman filter technologies for this EO/IR and radar application. As a result, multiple-wavelength signals and multiple phenomena are fused for improved discrimination and kill. As a future element of the NMD "system of systems," ORINCON's rapid, all-source data fusion enablers will offer the Missile Defense Agency enhanced discrimination capabilities. Anticipated Benefits/Commercial Applications: Adapting multiple-hypothesis fusion of multispectral data offers a means of achieving a substantial target discrimination capability at a reasonable cost and schedule risk. It would enhance BMD sensing capabilities for expanded Capability-1 ballistic threats and beyond, and greatly improve BMD surveillance (BMC4I) capabilities. The multispectral data fusion aspects of the proposed effort could benefit advanced medical imaging systems by providing a more accurate imaging capability.

Keywords:
Hyperspectral imaging, Multiple-hypothesis fusion engine, All-source data fusion, Hit-to-kill sensors, Target discrimination, BMC4I, Interceptor

ORINCON's Fusion-Generated Target Discrimination effort will develop innovative algorithms for Project Hercules and the Missile Defense Agency (MDA) that address complex NMD technological hurdles in the areas of discrimination, tracking, and decision support. Proven multiple-hypothesis fusion and extended Kalman filter algorithms provide super-resolution of target cloud objects by removing multiple-sensor pointing error biases. Elimination of these systematic pointing errors enables subsequent development of feature phenomenology algorithms to identify individual objects within the target cloud. Feature phenomenology algorithms, operating in real time, measure and extract key features such as object length change, object ballistic extension, and object spin rates. In this way, it is possible to discriminate among warhead(s), decoys, and penetration aids in short detect-to-kill times. To accumulate target cloud evidence, the current Project Hercules decision support architecture delays engagement decisions as long as possible. ORINCON proposes to investigate relaxation methodologies in a temporal Bayes net to develop an architecture to make faster, better engagement decisions. As future elements of the NMD "system of systems," ORINCON's bias error removal, multiple-hypothesis tracking algorithms, target discrimination algorithms, and decision support optimization will offer Project Hercules and the Missile Defense Agency enhanced target discrimination and trajectory prediction. Anticipated Benefits/Commercial Applications: The use of multiple sensors within a multiple-hypothesis tracking systems offers MDA a means of achieving substantial target discrimination and robust tracking capability at a reasonable cost and schedule risk. It would enhance NMD sensing capabilities for expanded Capability-1 ballistic threats and beyond, and greatly improve NMD surveillance (BMC4I) capabilities.

Keywords:
Target Discrimination, Multiple-Hypothesis Tracking, Target Object Map, Bias Estimation, Feature Phenomenology, Harmonic Structure Function, Hough Transform, Decision Support Optimization