Method could enable autonomous drones to monitor safety of bridges

Rust never sleeps, and cracking concrete doesn’t get a day off either.
The Jan. 28 collapse of Pittsburgh’s Fern Hollow Bridge was a dramatic reminder of that fact. The exact cause of the collapse won’t be known until the National Transportation Safety Board completes a months-long study, but Carnegie Mellon researchers have developed autonomous drone technology that someday might prevent similar catastrophes and lesser mishaps caused by deterioration.
Working with Shimizu Corp., a Tokyo-based construction and civil engineering company, CMU’s Robotics Institute built a prototype drone designed for monitoring bridges and other infrastructure. As part of that effort, researchers recently unveiled a new method that enables automated systems to more accurately detect and monitor cracks in reinforced concrete.
Sebastian Scherer , an associate research professor of robotics and leader of the CMU team working with Shimizu, said the crack-detection method was one of several technologies that the university developed for the project, which concluded in February 2022. The researchers built a working prototype of a bridge-monitoring drone that employs the crack-detection system and plan to use it at the Frick Park site of the Fern Hollow Bridge to make a detailed model for the post-collapse analysis.
“The automated technology we developed for the Shimizu project is designed to prevent this type of collapse via comprehensive mapping, crack detection and structural analysis that would be too much work if it were done by hand,” Scherer said. “Today, typically you only do spot checks on critical parts, since an exhaustive survey and analysis would be too slow. Automated defect-detection technology would enable inspectors to check bridges more frequently and perhaps identify problems before failures occur.”
Kris Kitani , an associate research professor of robotics, led the research team, whose system improves existing crack-detection algorithms by 10%.
Researchers in CMU’s Robotics Institute are using specially developed drones and new methods to detect cracks in concrete to better monitor and inspect infrastructure like bridges.
Using computer vision, CMU researchers created a new method to more accurately detect and monitor cracks.
Their system relies on a computer vision framework known as a convolutional neural network (CNN), a class of artificial neural networks commonly used to analyze visual imagery. This framework can readily identify animals and objects such as vehicles and production parts. But Jinhyung “David” Park, a senior in the Computer Science Department majoring in artificial intelligence and a member of Kitani’s research team, noted that these systems have much more difficulty detecting cracks.
“Cracks, unlike these other objects, are very thin – often only two or three pixels wide in images,” Park said. They also have extremely irregular shapes and can be obscured by marks or shadows on surfaces.
Even when existing systems detect a crack, they cannot always determine how wide it is. They often overestimate the size, which makes it difficult to accurately determine if the crack is a serious defect or whether it might be expanding.
The researchers addressed the problem by using reinforcement learning, a form of artificial intelligence. In this game-like approach, the computer uses trial and error to develop tactics for solving a problem while maximizing its performance based on rewards and penalties.
This approach is commonly used to improve the performance of robotic arms, Park said. In those cases, a robotic system uses reinforcement learning to analyze the space surrounding a robotic arm to determine which motions would best enable the arm to accomplish a goal.
As adapted by the CMU researchers for crack detection, reinforcement learning allows a computer system to analyze each pixel within an image. Rather than decide how to move a robotic arm, the system calculates the probability that each pixel is part of a crack.
“That might sound time-consuming because we’re considering every pixel,” Park said. “But our reinforcement learning agent is convolutional, so it does these predictions asynchronously for every single pixel at once.”
This technique enables the system to not only detect cracks at high resolution, but also to calculate the probability that separate cracks might be part of one larger crack.
“If you have a crack that’s kind of shaped like this” Park said, extending the thumb and pinky of his right hand, “and another one here” – extending his left thumb and pinky near his right hand – “but they’re disconnected in the middle, maybe the network can recognize, hey, these cracks are actually parts of the same crack.”
The team presented its crack-detection work in September at the IEEE International Conference on Image Processing (ICIP 2021) in Anchorage, Alaska, where it won the Best Industry Impact Award. In addition to Park, who was the lead author, Kitani’s team included Yi-Chun Chen, a Robotics Institute master’s student in computer vision; and Yu-Jhe Li, a Ph.D. candidate in the Robotics Institute and the Electrical and Computer Engineering Department.

Biden Delivers Passionate Call To Action To Invest in American Innovation

CMU Researchers Train Autonomous Drones Using Cross-Modal Simulated Data

Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, PA 15213
412’268 -2900


Diverse Origins of Earth’s Minerals: Helping Find New Minerals, Habitable Planets, and Extraterrestrial Life

A newly completed 15-year study details the origins and diversity of every known mineral on Earth. An innovative new perspective on mineralogy, findings rank amongst the field’s most important in the last century. Scientists decipher and catalog the diverse origins of Earth’s minerals; will inform models of life’s history, ...

View more: Diverse Origins of Earth’s Minerals: Helping Find New Minerals, Habitable Planets, and Extraterrestrial Life

How to find marker genes in cell clusters

Graphical abstract. Credit: Journal of Molecular Biology (2022). DOI: 10.1016/j.jmb.2022.167525 Which genes are specific for a certain cell type, i.e., “mark” their identity? With the increasing size of datasets nowadays, answering this question is often challenging. Often, marker genes are simply genes that have been found in specific cell ...

View more: How to find marker genes in cell clusters

Here are the most effective things you can do to fight climate change

The first three charts show, in descending order, how effective different behaviours are at cutting emissions. Credit: Max Callaghan, Author provided Limiting global warming to 1.5℃ above pre-industrial levels requires reaching net zero emissions by the middle of this century. This means that, in less than three decades, we ...

View more: Here are the most effective things you can do to fight climate change

How reindeer eyes transform in winter to give them twilight vision

Credit: Shutterstock Reindeer are loved the world over for their dark, expressive eyes, majestic antlers and magical association with Santa Claus. The moment you learn the cold, hard truth of how Christmas presents arrive under the tree is a harrowing one that blights many a childhood. But reindeer are ...

View more: How reindeer eyes transform in winter to give them twilight vision

Decrease in CO2 emissions during pandemic shutdown shows it is possible to reach Paris Agreement goals

Global daily emissions changes in 2019 and 2020. Credit: Nature Geoscience (2022). DOI: 10.1038/s41561-022-00965-8 An international team of researchers has found that the sudden drop in CO2 emissions during the early days of the COVID-19 pandemic demonstrates that it is possible to reduce emissions enough to meet the 1.5 ...

View more: Decrease in CO2 emissions during pandemic shutdown shows it is possible to reach Paris Agreement goals

Characterizing the materials for next-generation quantum computers with nonlinear optical spectroscopy

Imaginary part of the 2D spectrum of the Kitaev ring in (a) the topologically trivial phase with μ=0.005Λ, w=Δ=0.495Λ, and (b) the nontrivial phase with μ=0.495Λ and w=Δ=0.005Λ for N=60. Credit: Physical Review Letters (2022). DOI: 10.1103/PhysRevLett.129.017401 Researchers at the Department of Physics and the Cluster of Excellence “CUI: Advanced Imaging ...

View more: Characterizing the materials for next-generation quantum computers with nonlinear optical spectroscopy

Waiting for the sun to set to find a rare bird

A 4-day old whip-poor-will chick. Like their parents, developing chicks have cryptic coloration that allows them to go unseen on the ground.  Credit: Sarah Stewart When most people are just getting home from their workdays, I’m about to start mine. I am a researcher studying the breeding behavior of ...

View more: Waiting for the sun to set to find a rare bird

Research reveals the complexity of loneliness during lockdown

Credit: Unsplash/CC0 Public Domain A study exploring loneliness and solitude during lockdown highlights how different people experienced social restrictions as either “hell or bliss.” The study, published in Analyses of Social Issues and Public Policy and by researchers from the University of York, gathered stories of lockdown loneliness from ...

View more: Research reveals the complexity of loneliness during lockdown

Scientists unravel working principles of a promising material for fast-charging batteries

Lung cancer: When radiation does not work

Researchers find evidence for periodic disk instabilities in the massive nucleus of galaxy NGC 4258

Michio Kaku: “We need a second planet”

Speed-dating experiment reveals what makes a great first date

Fauci reports COVID rebound, says his is “much worse” than initial illness

Is Something Else Hidden at the Center of the Milky Way?

UCL releases animal research statistics alongside fellow top institutions

Walking Can Improve Cognitive Function

Rocket Report: ULA starts military lobbying campaign, SLS to launch in 2 months

The Race to Build Wind Farms That Float on the Open Sea

A Better Kind of Face Mask: Researchers Develop Virus Killing Masks


Top Car News Car News