Pittsburgh boasts 440 bridges — more than the canal-laced city of Venice, Italy. To keep travelers crossing the city's three rivers safely, these many bridges require a lot of maintenance, which can lead to massive maintenance and repair costs. As a result, sensors are being increasingly used to help keep better track of a bridge's structural health, and avoid major, costly repairs in the long run.

Now researchers at the University of Pittsburgh and the University of Notre Dame received a $500,000 award from the National Science Foundation to develop artificial intelligent (AI) technology that could help better monitor bridges and other structures, all while running on renewable energy.

Jingtong Hu, lead researcher on the project and associate professor of electrical and computer engineering at Pitt’s Swanson School of Engineering, says the project was motivated by the fact that most current, wireless sensors are powered by batteries. These kinds of devices are often placed in hard-to-reach areas, which can make maintenance or battery replacement challenging.

“Many times, it is costly, inconvenient, or even infeasible to replace or charge these batteries after deployment,” says Hu, adding that sensors and devices are installed under bridges, under water, or in other places that could present hazards for workers.

Hu and his team — who will collaborate with Yiyu Shi, associate professor of computer science and engineering at the University of Notre Dame — hope to extend the life of these structural sensors and devices by adding a complementary technology that acts as a smaller second sensor. The second sensor will run on renewable energy, including solar, wind, and thermal. In addition, an AI program will train the second sensor to switch the main sensor or device on only when it recognizes certain events or patterns, such as unusual vibrations. This will prevent the battery from being drained and needing to be replaced. 

“The main device is programmed to do all of the legwork,” says Hu. “The smaller sensor is the watchdog that can monitor the environment and wake up the larger sensor when necessary.”

He adds, “By applying AI, we hope to increase the lifespan of unattended sensors and make them more reliable and useful.”

The group has tested the technology in the lab but has yet to test it on an actual bridge.

Even so, Hu believes the technology could be applied to bridges in Pittsburgh and throughout Pennsylvania, where sensor has been built into places like the I-81 Wade Bridge in Dauphin County.

The researchers also see the technology being used in a variety of applications across the public and private sectors. This includes helping to predict natural disasters, as sensor technology is now used to observe gases emitted by active volcanoes, as outlined in a 2016 article from Smithsonian Magazine. Adding the second sensor could prevent volcanologists from having to check sensors placed in dangerous environments, where they can be exposed to extreme heat and dangerous gases.

If successful, the project may ultimately allow these devices to be powered by the environment, all while preserving the integrity of structures through safer, more sustainable methods.

“By applying AI, we hope to increase the lifespan of unattended sensors and make them more reliable and useful,” he said.