Description: We propose an innovation for a data-driven Artificial Intelligence/Machine Learning (AI/ML) model for uncrewed aviation fleet and flight operations optimization (UA-FFOpt) that will be trained with real-world data from an active flight program in the State of Michigan and at a partner cattle ranch in Florida. UA-FFOpt will leverage our existing Intellectual Property (IP), AVIATE, and NCR. AVIATE’s capabilities for optimized mission planning and NCR’s access to real-time weather and airspace constraints accelerate UA-FFOpt’s path to operation. Our model will incorporate aircraft capabilities across the fleet, airspace operational ‘rules of the sky’ for integration into airspace classes, and dynamic re-planning based on real-time events such as traffic encounters, varying weather conditions, and community and geographic constraints. UA-FFOpt will be developed to be transferrable to other fleet operations applications for other use cases, including larger uncrewed aircraft, destined for the AAM ecosystem. By its nature as an AI/ML-based approach, it will be extensible to incorporate additional ecosystem (e.g., Class B, C, and D airspace) and operational variables beyond our primary use case with additional data and training.

Benefits: Optimization scheme for broad mission, equipment, weather, and airspace constraints in AAM fleet operations. Requirements and methods for UAS Traffic Management/AAM operator fleet optimization using 1
Extensible model for multi-variable uncrewed aircraft (UA) operational fleet optimization. Framework for scalability of team (1