(7) Navy Technology Acceleration - Integration of Automatic Dependent Surveillance

Extracting patterns from Automatic Dependent Surveillance-Broadcast (ADS-B) data to identify air corridors and detect anomalous behaviors could provide crucial information for both commercial and military applications. Historically, pattern recognition and anomaly detection were dependent on statistical analysis. Patterns were defined as statistical models and anomalies were defined as outliers. Advancements in machine learning have allowed for identification of more complex patterns and adaptive anomaly detection. A method of machine learning known as online learning updates a neural net model in real time by continuously incorporating new data. This is essential for spatiotemporal data, which has confounding variables that alter patterns over time. Since flight routing is constantly changing due to temporal variables such as weather, identifying air corridors and detecting anomalous aircraft behavior requires an online approach that takes into account such variables. It is also feasible to develop a small, inexpensive, ADS-B anomaly detector that operates independently or in a network, on stationary and mobile platforms. In Phase I, Skyward will develop a strategy that will be implemented in Phase II to develop software and detectors to provide a flexible, scalable, and portable anomaly detection solution.