Autonomous Inspection of Solar Panels and Wind Turbines Using YOLOv8 with Quadrotor Drones

This paper introduces an autonomous inspection system for solar panels and wind turbines utilizing Tello drones and the YOLOv8 object detection algorithm. The main objective is to establish an efficient method for identifying and evaluating these renewable energy components' conditions, focusing on detecting issues such as breakage and dust accumulation. The system involves a pair of Tello drones operating as a swarm and connected to a standard router to enable real-time video streaming and data processing. The drones utilize the YOLOv8 algorithm for object detection, and Python programming is employed to manage their operations. The methodology encom-passes establishing reliable communication among the drones, router, and laptop, initializing the drones, capturing real-time video, and utilizing YOLOv8 for object recognition and clas-sification. The paper presents case studies demonstrating the system's effectiveness in detecting and classifying solar panels and wind turbines under varied conditions. While the system exhibits promise in reducing manual inspection labour and enhancing safety, limitations related to image quality suggest that using higher- resolution cameras could further improve its efficiency.