Air to ground object detection for airport runway – FOD clearance

• Main Author: Sundaresan, Sivasuriyan
• Other Authors: Mir Feroskhan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Aeronautical engineering
• Online Access: https://hdl.handle.net/10356/168349
• Institution: Nanyang Technological University

Foreign Object Debris (FOD) found on airport runways causes significant damage to the aviation industry, estimated to be up to USD 1.1 billion annually. FOD is also a high-risk factor that has led to major accidents such as Air France flight 4590. As the number of flights worldwide continues to increase, ensuring runway safety by effectively managing FOD is paramount. Current methods of detecting FOD pose several limitations and cannot keep pace with the increasing number of flights. Hence, this report proposes and investigates the use of air-to-ground object detection for FOD clearance. Specifically, the report will investigate the use of YOLOv5, SSD, Fast R-CNN, and Canny edge detection algorithms on a dynamic system to detect FOD. The algorithms are evaluated based on their accuracy of detection, training speed, data set size, inference speed, flexibility of algorithm, size of object detectable, and robustness. Based on the evaluation, Canny edge detection is chosen, and to overcome its shortcoming, the report suggests using object size detection along with the Canny edge detection algorithm. The report investigates the optimal flight parameters for drones through flight tests and proposes a FOD detection ecosystem that airports can implement. From this report the potential benefits of air-to-ground object detection for FOD clearance, such as increased accuracy and efficiency are known, which could lead to a safer and more cost-effective FOD management system.