Hardware Identification & Distance Estimation for an High-Voltage Transmission Line Inspection Robot

Abstract

During the walking process of a patrol robot, it is necessary to recognize and measure the distance of hardware along the line. For high voltage transmission lines, a visual navigation method based on embedded system is proposed to achieve hardware positioning and recognition. Firstly, the template image is collected, and then processed and enhanced. Next, the edge or shape features of the hardware are constructed and extracted, which are then matched with real-time images to complete the identification of hardware on the line. By using the curve fitting of the hardware area and distance obtained from the principle of pinhole imaging, real-time and accurate estimation of hardware distance is achieved. The experimental verification was conducted on a simulated transmission line in the laboratory, and it was found that the identification system built in this paper could effectively recognize hardware on the transmission line and meet the design requirements during the robot's walking and patrol inspection process.