Air Traffic Control Systems
The safe and efficient flow of aircraft in today's skies depends on a suite of weather sensing and traffic management technologies. Our group aims to improve air transportation by developing sensors, weather tracking and forecasting systems, and decision support automation that assists pilots and air traffic controllers in routing aircraft out of harm's way and maintaining smooth traffic patterns. To ensure user acceptance and the effectiveness of our technologies, we rely on extensive analysis and field evaluations and draw upon diverse areas of expertise, including signal processing, meteorology, software engineering, machine learning, human factors, and more. The systems our researchers have developed currently operate nationwide to minimize weather-related delays and protect aircraft against accidents. In addition to air traffic management, we are conducting initiatives in transportation logistics, the reduction of aircraft noise and environmental impacts, and cyber-secure information systems.
Advancing Our Research
A deep learning-based velocity dealiasing algorithm derived from the WSR-88D open radar product generator
Foundational AI Concepts
Our group applies artificial intelligence (AI) and machine learning (ML) techniques to improve air traffic safety and collision avoidance systems. The video linked below provides an overview of how AI/ML capabilities are incorporated into operational workflows and describes several applications of this technology.
Transforming Cameras into Weather Sensors
The Visibility Estimation through Image Analytics (VEIA) technology provides an inexpensive and robust way to extract meteorological visibility from cameras – thus transforming cameras into weather sensors. The VEIA algorithm uses the presence and strength of edges in an image to provide an estimation of the meteorological visibility within the scene. The algorithm compares the overall edge strength of the current image to those from a clear day representation generated by creating a composite image from several days of imagery.