SBIR-STTR Award

Complex sensor systems available on Naval Unmanned Air System (UAS) platforms requires advanced techniques to enhance their resiliency and survivability. This includes autonomous artificial intelligence (AI) architectures that can process, analyze, and provide actionable intelligence in terms of understanding and reporting on adversarial actions/events focused on damaging/crippling Navy UAS system. This includes system failures, either through degradation over time or operational mistakes. Unmanned vehicles must operate in unstructured environments that are inherently unpredictable and dynamical. An autonomous UAS must have some degree of cognitive intelligence in order to undertake tasks without direct and continuous human involvement, especially in unknown environments. Colorado Engineering Inc. and ISEA TEK LLC, the CEI Team, proposes research into AI-enabled cognitive machine learning system for Real-time Autonomous Sensor Processing (RASP) technologies that will provide Unmanned Service Systems (USS) and Unmanned Air Systems (UAS) with the capabilities required for semi-supervised and autonomous command and control. This research will create the RASP system, utilizing an intelligent multi-agent processing infrastructure that will meet the Navy’s needs now and into the future. The focus of this research is to create AI-enabled technologies that provide complete sensor and technology integration for effective UAS mission.

The high-level objective for this Phase II research and development effort is to complete the development of the robust, adaptive, Multi-Agent Learning Systems for Cognitive Autonomous Sensor Processing (CASP) architecture and demonstrate the capabilities for unmanned, autonomous operation of UASs. This will provide the capabilities for supporting continuous, adaptive learning, and mission management decision making support to provide timely, accurate sensor collection and processing for the Navy system integration and execution. This will include a prototype of the CASP Artificial Cognitive Neural Framework (ACNF) cognitive processing infrastructure required to manage and execute the CASP cognitive system. We will advance sensor processing algorithms developed in Phase I to encompass several more sensor types and algorithm types, providing a sensor fusion suite of tools applicable to a variety of Naval missions and operations. The Phase II demonstration will provide a proof-of-concept for the CASP ACNF cognitive framework. This will include demonstration of utilities and sensor fusion capabilities required for a specified program of record.