The rise of small unmanned aerial systems (sUAS) near airports presents growing safety risks, including mid-air collisions, operational disruptions, and security threats. Current detection systems, such as radar and optical tracking, struggle to reliably identify and classify aerial — particularly non-cooperative drones — under variable operational conditions.
This research proposes a multi-modal aerial object detection system that combines six sensor modalities to enable real-time surveillance. By integrating sensor fusion and machine learning (ML), the system aims to improve detection and classification accuracy, reduce false positives, and support Real-time Decision-making for airport safety personnel. The research aims to evaluate system performance under varied weather and lighting conditions and develop a scalable framework for enhanced Situational Awareness in airport operations.