UWB and GNSS Hybrid and Cooperative Localization enabling UAV Bridge Inspection

In recent years, Unmanned Aerial Vehicles (UAVs) have emerged as a promising solution to perform critical tasks in a vast range of industrial applications and for monitoring large critical infrastructures. Bridge inspection, as one of them, is essential to ensure the structural integrity and safety of these vital infrastructures. By exploiting the advantages of the UAVs, it is possible to create a more safe and automated solution compared to the traditional inspection methods that often involve manual labor, time-consuming processes and potential risks to human inspectors. In this context, one of the most challenging problems is represented by the localization of the Inspector UAV beneath the bridge because of the limited or completely absent satellite visibility. This thesis proposes a solution to this problem by introducing 4 external UAVs, defined Supportive UAVs, with good GNSS visibility that move laterally along the bridge and at the same time can localize the Inspector UAV using Ultra-wideband (UWB) wireless technologies. Thanks to its signal structure, the UWB provides enhanced ranging accuracy, resistance against multipath effects and high time resolution, making it a reliable localization solution in GNSS-denied areas. The solution was first implemented and tested in simulation, readjusting an existing peer-to-peer positioning simulation tool to the specific scenario settings. Furthermore, some tests were performed in different steps: firstly an indoor localization of a UAV was carried out, taking advantage of VICON, a motion capture system based on infrared, made available by LINKS Foundation; then some indoor localization tests on the UAV with UWB were conducted, using VICON as ground truth reference, in a limited environment to simulate a GNSS-denied scenario. Finally, performance analysis and considerations on future work have been provided.