Aerial Localization and Navigation for Surveillance of Large, Featureless, GNSS‐Denied Maritime Environments

ABSTRACT In this study, we tackle the problem of surveilling a large area over the sea without access to the global navigation satellite system (GNSS) while searching for the known target vessel. The system utilizes an unmanned aerial vehicle (UAV) equipped with long‐range radio frequency (LoRa) localization and an RGB camera to detect the target vessel and reconstruct its position. To address issues such as low‐frequency updates and noisy LoRa data, we utilized a Kalman filter, while Nyquist analysis was used to ensure control stability under latency. We also demonstrate the UAV's ability to identify targets and reconstruct their pose while operating nearly a kilometer away from the home location. Finally, we present the results of the complete task. This includes precise takeoff and landing guided by visual‐based positioning using the AR tag markers. The UAV then transitions to the LoRa localization frame, surveys the area along predefined waypoints, identifies the target vessel, and successfully returns to the home location.