Sensor Data Fusion

Under this SBIR, the Decisive Analytics Corporation (DAC) team addresses two niche capabilities that are not currently addressed in the Ballistic Missile Defense System (BMDS). We propose describing the problems of tracking and discrimination within a single, hybrid, dynamic Bayesian network model. We propose solving this network model using a novel inference algorithm suited for use with hybrid dynamic Bayesian network. Additionally, we propose to address the problem of track correlation by employing an innovative Multiple Hypothesis Bayesian Network framework that works in conjunction with our proposed integrated tracking and discrimination network model. Both of these proposed approaches will build on the DAC team’s computational techniques for performing inference in large-scale network models. These technologies, combined with DAC’s experience in missile defense will result in algorithms suitable for prototyping in Phase I and Phase II, and integration and deployment to the BMDS via in Phase III of this SBIR.