ATAC — National Aeronautics and Space Administration SBIR Phase I: A3

Description

ATAC proposes ldquo;Traffic Pattern Integration for Autonomous Vehicles,rdquo; in response to Subtopic A3.02. Our proposal fulfills the goals of Subtopic A3.02 by creating a cyber-resilient, service-based architecture that can be used to integrate autonomous vehicles with legacy users around small towered and non-towered airports. A significant challenge for an autonomous vehicle flying into a small airport is the lack of defined landing procedures.nbsp; Most small airports do not have GPS procedures and instead rely on general rules for organizing traffic into a landing pattern around the airport; this is the case at both towered and nontowered airports. Our Traffic Pattern Intent Prediction (TPIP) tool overcomes this challenge by learning the traffic patterns at small airports and using them to predict aircraft intent in the traffic pattern around small airports. TPIP accomplishes this using an innovative approach that predicts the future intent of aircraft entering the traffic pattern using probability models trained on historical data. TPIP will help autonomous vehicles plan their route into an airport, avoid unsafe situations, and increase operational efficiency. The information provided by TPIP allows autonomous vehicles to strategically adjust their routes into crowded airspace in order to minimize the need for tactical maneuvers. It also provides dynamic route planning for landing trajectories that conform with local traffic patterns and avoid conflicts with other aircraft. This proposal fulfills the goals of Subtopic A3.02 by creating a service-based architecture that can be used to integrate autonomous vehicles with legacy users. Our solution will help safely scale autonomous operations while maintaining a safe and usable environment for legacy airspace users. At the end of Phase 1 we will create a proof-of-concept demonstration to show our ability to execute on these goals, as well as assess the accuracy of our model to ensure that the predictions are reliable.