Artificial Intelligence (AI) Algorithms for Sense and Avoid Neural Network Architectures for Orbs and UAVs

Previous attempts to integrate neural networks, artificial intelligence (AI,), sensors, and Orbs in the Urban Air Mobility (UAM) market have not been successful. Prior approaches have been impractical to the needs of the Air Force (AF) customer and end-user, and have not validated the product-market fit between prior solutions, AF stakeholders, and commercial UAM markets. VISIMO's novel approach will provide a dynamic decision and risk prediction engine for Orbs and Unmanned Aerial Vehicles (UAVs), complemented by innovative technical approaches based on artificial neural networks that have been tested and successfully applied in other industries and markets. A successful neural network approach, coupled with the design of AI and ML algorithms that can perceive, learn, decide, and act on their own, applied to disaster response, humanitarian aid, and logistics missions at scale, would advance the state-of-the-art (SOTA) in sense and avoid architectures in the context of Orbs/UAVs and the UAM market. The technical Phase I deliverables will validate the product-market fit between algorithmic and neural network sense and avoid architecture solutions and potential AF stakeholders through a feasibility study, as well as define a clear and immediately action plan to demonstrate value and mitigate potential risk proposed by the adoption of new algorithms within the proposed AF customer base. VISIMO will use the results of the feasibility study and the risk mitigation analysis as a way to inform the design of prototype algorithms and integrate neural networks with sensors for avoidance architectures and Orbs/UAVs within the Department of Defense (DoD), the AF, and the commercial UAM market.