“To do their job, the drone swarms must be precisely positioned in relation to the turbine. So, even under complex operating conditions, it is essential that we always know what the distance and angle are between the drone and the respective component of the turbine. This allows the drones to collect comprehensive data,” says Farzad Tashtarian (Department of Information Technology), who co-leads the project with Jan Steinbrener (Department of Smart Systems Technologies, Control of Networked Systems) at the University of Klagenfurt.

If a turbine develops defects, the drone swarms must be able to detect them. To this end, researchers are working towards implementing AI-supported image analysis that can detect even the smallest defects. To collect the data, drones are equipped with high-resolution cameras and sensors such as LiDAR, radar, and ultrasonic sensors.

The research team is pursuing several goals with DORBINE: It should no longer be necessary to shut down wind turbines for inspections. Robust and AI-based navigation systems should enable precise positioning of the drone swarms and ensure efficient data collection. Advanced AI and machine learning techniques should make it possible to accurately detect cracks, erosion and structural deformations in the recorded images, optimizing inspection plans and maintenance measures. Farzad Tashtarian adds: “To date, wind turbines have been inspected using helicopters and work boats – which generates significant CO2 emissions. By using drone swarms, we can achieve significant reductions in emissions.”

DORBINE’s project partner is AIR6 SYSTEMS. The project is funded by the Austrian Research Promotion Agency (FFG).