Sensor Data Fusion

The MDA mission requires the use of data fusion to obtain accurate and persistent tracks of threatening objects discriminated from debris/countermeasures in order to provide an integrated picture of the battle space as well as fire control solutions to interceptors. A wide variety of information may be available for integration, including multiple sensor measurements, features, and track states; combining all these can improve track identification, association and accuracy, especially in the challenging MDA environment. Toyon Research Corporation proposes a dual-layer solution for data fusion for both short-term individual measurement associations and longer-term track-to-target associations. A Multiple Hypothesis Tracker (MHT) will be used in conjunction with a Bayesian network to model feature information and possible inferences in a way that promotes improved measurement-to-track association. Toyon"s Fusion and Correlation for Tracked Object Retention (FACTOR) system will handle track fusion and feature database management/track stitching algorithms. The FACTOR system has been tested and validated in other environments and will be adapted for the MDA picture, and a semi-supervised learning algorithm to help fuse features will be added. Toyon will design a test scenario with multiple short- and medium-range attacks from multiple launch sites in order to test these algorithms.