Contextual Reasoning for Object Identification

This Small Business Technology Transfer project will develop a mathematical framework and enabling technologies for context-aware target classification and selection for missile defense systems in the presence of counter-measures and clutter in realistic scenarios. Based on statistical relationship learning (SRL) methodology, the proposed framework can reason simultaneously about multiple entries in the threat complex even in the presence of corrupted or missing sensory data. The Phase-I effort will include; development of context-based classification framework based on SRL, quantitative and qualitative evaluation of the proposed technologies, and demonstration of proof of concept using simulated data from multiple use-cases. The project will benefit from Novateur Teams expertise in the areas of radar systems, multi-sensor data and decision fusion, adaptive context-aware system, distributed sensor network and estimation theory. Approved for Public Release 16-MDA-8620 (1 April 16)