Commissioning an All-Sky Infrared Camera Array for Detection of Airborne Objects

Area of Science:

• Astronomy and Astrophysics
• Aerospace Engineering
• Computer Vision

Background:

• Limited scientific data exists on unidentified aerial phenomena (UAP).
• The Galileo Project aims to address this gap with a ground-based observatory.
• Existing UAP studies lack rigorous, long-term, multi-modal data collection.

Purpose of the Study:

• To establish a baseline performance for a new UAP detection system.
• To collect and analyze multi-spectral data of aerial phenomena.
• To develop robust methods for identifying and tracking UAP.

Main Methods:

• Utilized an all-sky infrared camera array (FLIR Boson 640).
• Implemented extrinsic calibration using Automatic Dependent Surveillance-Broadcast (ADS-B) data.
• Employed You Only Look Once (YOLO) and Simple Online and Realtime Tracking (SORT) for object detection and trajectory reconstruction.

Main Results:

• Achieved a 41% acceptance rate for ADS-B-equipped aircraft.
• Reported a 36% mean frame-by-frame detection efficiency, influenced by weather and range.
• Reconstructed approximately 500,000 aerial object trajectories over five months.
• Identified 144 ambiguous trajectories requiring further investigation.

Conclusions:

• The developed system provides a foundational dataset for UAP studies.
• Detection efficiency is significantly impacted by environmental factors and target characteristics.
• Further sensor modalities and kinematic data are needed for definitive UAP identification.