ISSN: 1304-7191 | E-ISSN: 1304-7205
Inspection of power transmission line insulators with autonomous quadcopter and SSD network
1Department of Mechanical Engineering, MNNIT Allahabad, Prayagraj, U.P., 211004, India
Sigma J Eng Nat Sci 2024; 42(3): 621-632 DOI: 10.14744/sigma.2024.00059
Full Text PDF


In the next generation of smart cities, Unmanned Aerial Vehicles (UAV) also known as drones are playing a vital role in many advanced applications such as power transmission line in-spection, transportation, aerospace and surveillance etc. Due to the excessively high and wide transmission tower heights, the conventional methods of power line inspection are general-ly ineffective. This manuscript’s primary focus is the development of an autonomous UAV/quadcopter that can hover over transmission towers and capture photographs and videos by flying along pre-planned routes. Quadcopters have a distinct feature that distinguishes them with the existing aerial vehicles and have a vital role in wide range of applications such as live monitoring of traffic and crowded areas, remote locations, delivery and inspection. This man-uscript also explains about the advanced sensors & components such as Global Navigation Satellite System (GNSS), optical flow sensor and Here Link etc. required for fabrication of an autonomous quadcopter for power transmission line applications. The fabricated quadcopter includes a light weight S-500 frame equipped with intelligent controller such as Pixhawk cube orange (2.1) and NVIDIA nano board for receiving and analyzing the data from the onboard sensors and camera based on pre-determined criteria. The proposed approach increases effec-tiveness and accuracy, has a promising future for intelligent insulator detection and inspection which is a valuable addition to power networks. The suggested deep learning technique has a detection speed of 51.8 frames/sec and a detection accuracy of up to 90.31 percent. The suggested DL algorithm has a promising future in terms of intelligent insulator inspection in power grids.