Artificial Intelligence Capabilities for Aviation Behavior Characterization and Anomaly Detection SBIR
The Department of Defense’s (DoD’s) Artificial Intelligence Strategy directs DoD to accelerate adoption of AI and creation of a force fit for our time . Realizing this vision requires identifying appropriate use-cases for AI across DoD, rapidly piloting solutions, and scaling use across the DoD enterprise. To support the DoD AI strategy, the Navy wants to develop AI/Machine Learning (ML) technologies to address a variety of use-cases. One of these use-cases requires the use of AI/ML technologies to autonomously characterize the behavior of self-reporting aircraft using Automatic Dependent Surveillance-Broadcast (ADS-B) data.
ATAC is developing a Human-Interpretable AI Learning System (HAILS) that is encapsulated as a software service with aircraft behavior models and data to be used to (1) identify apparent air corridors and (2) detect anomalous behavior in support of determining aircraft intent. Below is an example of air corridors automatically determined by ATAC’s AI/ML processes.