Wingbeat Over Wind Turbines: Autonomous Drones for Acoustic Bat Detection in Operational Wind Farms

The increasing number of wind turbines has a negative ecological impact on bats. To improve real-time acoustic bat monitoring at wind farms, this study investigated the technical requirements, options, and challenges of operating autonomous unmanned aerial systems (UAS). The use case required autonomous landing and recharging, obstacle avoidance, and acoustic noise mitigation. A real-world test suggested that single-anchor ultra-wideband ranging provided sufficient accuracy for landing, radar odometry to avoid wind turbine blades required further research, and UAS ultrasonic noise emissions overlapped with bat echolocation calls. For noise mitigation, this study found that a bat detector placed 10 m below the source did not require acoustic attenuation, but the system benefited from deep learning methods to remove the remaining noise. Our results demonstrate the current state of the art and potential research opportunities for using UAS to improve smart curtailment strategies in wind turbines to balance renewable energy production with wildlife conservation.