Visual-based Assistive Method for UAV Power Line Inspection and Landing

Over the years, many methods and technologies have been developed to improve power lines’ visual-based inspection, such as the recent advances in computer vision and machine learning techniques and the use of Unmanned Aerial Vehicles (UAVs). Although aerial inspection of transmission lines is a well-researched topic, there is still space for research papers and solutions that focus on shared control, landing, and experimental evaluation of the solutions. This kind of system has the capabilities to acquire and process information about its surrounding. By employing automatic UAVs embedded with artificial intelligence, industries, business and researchers can significantly improve their visual-based inspection routines, bringing safety to the user and processing reliable information. Therefore, this research work proposes a strategy to both detect and track power transmission lines and a method to allow assistive control during UAV landing. Both methods were evaluated in simulated and real-world scenarios. Regarding the detection and tracking strategies, the outcomes suggested that the proposed system is capable of correctly identifying power transmission lines and navigating above them, even in the presence of cluttered backgrounds. Futhermore, the results of the assisted landing strategy showed that the method has excellent performance and is technically viable for practical deployment.