Bayesian Tracking Algorithm(s) for Determining Highest Probability Predicted Intercept Points(s) in the AEGIS Combat System

This Phase I project will demonstrate the use of a Bayesian-based tracker can be used to instantaneously addressing targets in raiding or swarming configurations and provide optimal engagement options to the Sailor. The proposed approach is to develop the tracking and engagement scheduling algorithms using customized tools for dynamic Bayesian Networks and Decision Influence Diagrams within a Probabilistic Relational Model framework. The tools will include the modeling of Second Order Uncertainty (SOU) in capturing and quantifying uncertainty in the provided target observations and the conditional probability processing elements used by the Bayesian-based tracker. The desired outcome of our Phase I Option effort will be a successful demonstration of the our SOU technology using a toy prototype simulation that identifies the highest probability Predicted Intercept Points with a threat scenario. The simulation can be exercised to demonstrate the interaction of the various relevant system elements (threat types, assumed capability of the incoming threat, number of threats, operational and test environment conditions, clutter and debris) that have the highest probability of kill. This includes establishing the confidence/uncertainty levels when predicting performance of the modelled system