Master Thesis

Neural point features for 3D Drone Localization inside Wind Turbine Blades

Work default illustration


  • Started: 01/02/2023
  • Finished: 17/07/2023


Internal drone inspections have recently appeared as an efficient solution for detecting wind turbine defects that are externally undetectable. The goal is to develop deep-learning features inspired by classic descriptors such as SIFT or SURF that are exploited in SLAM techniques to map and localize where the drone is flying inside the wind turbine blade. Localizing the blade defects would help in designing a proper blade repair recommendation. This industrial project will be executed in collaboration with the company Wind Power Lab, who will provide the video data.