Automatic Deep Learning-based Target Identification with Semantic Reasoner for Autonomous Collaborative Weapons

To address the Air Force’s need for target identification schemas for autonomous collaborative weapons systems, Physical Optics Corporation (POC) proposes to develop a new Automatic Deep Learning-based Target Identification with a Semantic Reasoner for Autonomous Collaborative Weapons (ATRACE) software suite that enables each weapon in a collaborative weapons team to detect and identify targets. It is based on a new design that leverages POC’s existing technologies. Specifically, the innovation in semantic reasoning, deep learning-based target detection and identification, and real-time reduction of high?dimensional data streams into abstracted information using low?size, weight, and power (SWaP) hardware will enable the ATRACE software to accurately detect and identify targets for collaborative weapons systems. In Phase I, POC will demonstrate the feasibility of ATRACE by developing a proof-of-concept prototype using POC’s existing datasets, reaching technology readiness level (TRL)-3. In Phase II, POC plans to develop and demonstrate an ATRACE prototype software suite, reaching TRL-5/-6.