Bird protection at wind turbines using 3D tracking

The energy transition in Germany is aiming for an 80 % share of renewable energies by 2030. Similar goals exist in other European countries. Wind power generation plays a key role in this, but this often conflicts with nature conservation. Protected bird species in particular must be protected from collisions. The bird detection systems available on the market often do not meet the requirements of the wind farm operators. The aim is to develop a precise measuring system that can track birds over long distances in 3D. This is to be achieved using a novel combination of static wide-angle cameras and gimbal-mounted telephoto cameras with laser distance measurement modules. The measuring system shall detect birds at large distances and record the 3D flight trajectory in high resolution. The collision risk can be derived from this data with a high degree of probability. The shutdown of wind turbines can thus be limited to cases where there is a specific risk to a bird. This saves considerable costs and reduces wear and tear. We present results achieved with a prototype of the measurement system. Detection and tracking of birds in the camera images are done using a combination of performance-optimized heuristic algorithms and machine learning models. Individual birds can be tracked over distances greater than 250m and with a spatial resolution of at least half a meter in all spatial directions.