Sign Up for the Newsletter
The most up-to-date news and insights into the latest emerging technologies ... delivered right to your inbox!
The system combines machine learning, lidar and digital twin technologies to monitor and assess structural damage in urban infrastructure
February 1, 2024
Researchers have developed an AI-enabled system to power robotic inspections of buildings and infrastructure.
The team, from Drexel University in Philadelphia, Pennsylvania, created the system in response to the rapidly aging built environment, with the rise in structural failures of buildings, roads and bridges.
Automating the inspection process could help target the problem before it worsens and the new AI-powered system is designed to speed up inspections and maintenance.
Deployed in autonomous robots, the system combines computer vision with a deep-learning algorithm to pinpoint problem areas, using lidar to create a series of laser scans of the regions and generate a digital twin to assess and monitor the damage.
The system takes the same program model as that used for facial recognition, drug development and deepfake detection – selected for its ability to spot precise details in vast amounts of data.
The program was trained on a dataset of sample cracks and can identify imperfections in the images that the robotic system collects from the surface of a concrete structure.
"Cracks can be regarded as a patient's medical symptoms that should be screened in the early stages," the authors wrote. "Consequently, early and accurate detection and measurement of cracks are essential for timely diagnosis, maintenance, and repair efforts, preventing further deterioration and mitigating potential hazards."
Another issue the team hopes to address is the ongoing labor shortage of skilled infrastructure workers, who inspect and repair these aging structures.
"Civil infrastructures include large-scale structures and bridges, but their defects are often small in scale," said Arvin Ebrahimkhanlou, study author. "We believe taking a multi-scale robotic approach will enable efficient pre-screening of problem areas via computer vision and precise robotic scanning of defects using nondestructive, laser-based scans."
The team noted that human inspection workers would make the final call on when and how to repair damages, with the robotic assistants just providing a helping hand in lightening inspectors’ workloads.
Looking forward, the team hopes to incorporate the system into a larger autonomous monitoring framework that includes drones and other autonomous vehicles – such as the one proposed by the Federal Highway Administration's Nondestructive Evaluation Laboratory, which would use an array of tools and sensing technologies to autonomously monitor and repair infrastructure.
"Moving forward, we aim to integrate this work with an unmanned ground vehicle, enhancing the system's ability to autonomously detect, analyze, and monitor cracks," said Ali Ghadimzadeh Alamdari, a research assistant. "The goal is to create a more comprehensive, intelligent and efficient system for maintaining structural integrity across various types of infrastructure,
Additionally, real-world testing and collaboration with industry and regulatory bodies will be critical for practical application and continuous improvement of the technology."
Assistant Editor, IoT World Today
Scarlett Evans is the assistant editor for IoT World Today, with a particular focus on robotics and smart city technologies. Scarlett has previous experience in minerals and resources with Mine Australia, Mine Technology and Power Technology. She joined Informa in April 2022.
You May Also Like