International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 12 | Dec 2024
p-ISSN: 2395-0072
www.irjet.net
Autonomous Utility Pole Identification David Pesin1, Elizabeth Piersall1 1Oak Ridge National Laboratory, One Bethel Valley Rd., Oak Ridge, Tennessee, 37831 USA
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract – The implementation of small unmanned aerial
following a detected fault. The goal is to develop a functional prototypical software that automatically parses a video from an sUAS overflight with its respective flight log and outputs a set of images tagged with all relevant information including GPS and attitude information, a timestamp, and a pole identification number if there is a utility pole present in the frame. This not only makes it immediately easier to use this data on its own, but also significantly enhances the ability to train a machine learning model off it. The system can be separated into four major components: data acquisition, data extraction and encoding, data filtration, and pole tagging.
systems (sUAS) for the purpose of powerline inspection is an emerging concept among utility companies. Such operations can produce a significant amount of useful data for the purpose of training different machine learning models or keep track of the integrity of electrical infrastructure. As such, this system is developed for the purpose of properly cataloging and leveraging this collected data. Key Words: Powerline inspection, unmanned small aerial systems, machine learning
Data acquisition broadly encompasses any overflight of a pole route that collects video footage. For our purposes, we are using platforms that leverage the open-source autopilot ArduPilot which can be run on a variety of flight controllers and multiple different sUAS form-factors such as multirotor drones, fixed-wing aircraft, and even blimps. Regardless of form-factor this software allows for the simple autonomous extraction of the logfiles after flights. In this case, the ArduCopter derivative is used. These sUAS also allow for simple video recordings of our entire flight to be saved as an MP4. Because we are using the same drones and cameras for all acquisition, the formatting of the videos and data is the same.
1.INTRODUCTION Using an sUAS has become an increasingly popular method for the purpose of patrolling power systems and routine visual inspections of utility infrastructure. [1] Equipped with a camera, the pilot can inspect a circuit manually as the flight is occurring. Most platforms with such functionality also offer the ability to record footage of the flight. In this case, an operator can inspect the footage after the fact, but that footage has far more uses than just retrospective analysis by an individual. Such footage can be used to train any machine learning model for the purpose of object recognition and component analysis, something that has significant utility and interest.[2] The difficulty is converting the video into a useful and usable format for such a task.
After the data is collected and extracted from the sUAS, video and telemetry information are parsed separately then combined. The video itself is split into frames using the Python package PyAV, while the logfile is parsed with assistance from PyMavlink. The extraction system can take any uploaded .MOV or .MP4 video file and separate it into either every frame or just the keyframes whilst simultaneously storing the uploaded logfile in memory after converting it into a usable format and deleting irrelevant information. This information is then processed together, matching each frame to a corresponding line in the logfile with the nearest timestamp. That logline is then parsed and linked to the frame itself. This is done by either using the Python package PyExiv2 to tag the EXIF metadata of each frame with relevant information, or by simply storing each frame index and relevant log line in a CSV file for future database storage depending on the utility’s needs. The data is then filtered for the purpose of removing unhelpful data via an altitude threshold and a line detection algorithm.
Fig-1: A compromised utility pole where the wood has rotted due to damage caused by burrowing woodpeckers. The intent of this project was to lay the foundation for enhancing powerline inspection routines by creating a system for the future automation of initial circuit patrols
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 321