AI robot enhances infrastructure inspection
Cities worldwide consist of roads and buildings that withstood decades, but they won’t stay for long. Experts believe the world faces an aging infrastructure problem, but we aren’t doing enough to solve it. Soon, our lack of action may lead to the loss of lives and billions of dollars in repairs and reconstruction.
That is why Drexel University researchers developed a new system that uses robots and artificial intelligence to check buildings. Specifically, they detect and asses structural, such as cracks that human inspectors can’t find themselves. Soon, it could become a staple for repairing structures worldwide.
How does this AI robot work?
.@DrexelEngr researchers are using AI to make it easier to find and monitor problematic cracks in concretehttps://t.co/cV6TrFfiwZ pic.twitter.com/kVKAurR9mH
— Drexel U. News (@DrexelNews) January 30, 2024
Drexel’s AI robot consists of a computer vision module and a laser scanning module. It uses a deep-learning algorithm called a convolutional neural network to analyze high-resolution image structures and spot cracks.
Then, the laser scanner performs a deep analysis of the cracked regions and creates a 3D damage model. Later, experts could use it for further monitoring.
The researchers compare cracks to medical symptoms that require early diagnosis and treatment to prevent more serious issues. Consequently, their AI robotics system can facilitate inspection to provide more focused and timely interventions.
Interesting Engineering says the world’s bridges, roads, buildings, and tunnels show signs of decay and damage. Soon, they could collapse unless we repair them immediately.
“Cracks are the first signs of structural deterioration and potential hazards that should be detected and measured as soon as possible,” said researcher Arvin Ebrahimkhanlou, Ph.D.
READ: Scientists develop self-healing concrete
“Our system can provide a fast and accurate way of screening and scanning cracks, which can help with diagnosis, maintenance, and repair efforts,” he added.
The researchers emphasized infrastructure requires a multi-scale robotic approach that can handle small and large defects. “Civil infrastructures are composed of massive structures and bridges, but their problems are often microscopic,” Ebrahimkhanlou stated.
“We believe that our system can offer a solution that can combine the advantages of computer vision and laser scanning, and enable autonomous robots to perform efficient and precise inspection of defects.” Learn more about this research on ScienceDirect.
How does AI help the homeless?
Societies must ensure citizens have adequate housing. That is why the University of California’s Los Angeles California Policy Lab (CPL) created an AI tool that detects people likely to lose their homes.
It analyzes over 400 types of records, including emergency room visits, receipts of public benefits, arrests, and other interactions with local systems. Then, the CPL algorithm accurately predicts who is at risk for homelessness within the next 12 months.
Moreover, homelessness prevention project personnel receive a list of medical records every three months. They initially keep them anonymous and then match them with corresponding individuals.
That allows staff to cold call them. Dana Vanderford, the associate director of homelessness prevention with the Los Angeles County Department of Health Services, admits these records should be private and personal.
Yet, Vanderford said they must use this data to help find those who may lose their houses. She says many don’t reach out to the many referral support programs in Los Angeles because they don’t know where to start.
READ: Russian cosmonauts find cracks in ISS module
Some don’t realize the risk of their living situation. Also, the associate director explained, “We have clients who have understandable mistrust of systems. [They] experienced generational trauma. Our clients are extremely unlikely to reach out for help.”
Vanderford and her colleagues have aided 560 residents by cold calling those identified by the CPL algorithm. “I believe deeply that there’s a need for targeted prevention programming,” Vanderford stated. “Without the ability to use AI, we don’t have a good shot at targeting these resources to those who need them most.”
The homelessness prevention program provides individual clients $4,000 or $6,000 of support, while families get $6,000 or $8,000 US plus additional funds, depending on the number of people in the household. Learn more about this project in my other article.
