Researchers from the University of Pittsburgh and the University of Notre Dame are developing a new method to extend the lifetime of sensors and devices deployed in remote areas.
Remote sensors, which can be fastened to a bridge to detect vibrations or produce data for structural health monitoring, require frequent maintenance. The problem is that these sensors are installed in hard-to-reach areas, complicating the maintenance process and potentially putting workers in harm’s way.
"One of the major challenges with these sensors is battery replacement,” said Jingtong Hu, associate professor of electrical and computer engineering at Pitt's Swanson School of Engineering. “Many times, it is costly, inconvenient or even infeasible to replace or charge these batteries after deployment."
The team developed a way to save power on remote sensors by using energy harvesting technology powered by solar, thermal or wind. The plan is to add a second, small sensor that can trigger a more robust device, saving more energy and allowing users to change the battery even less frequently. With the help of artificial intelligence, the sensor can be trained to recognize patterns and signal the larger device to turn on during a specific event.
"The main device is programmed to do all of the legwork," Hu said. "The smaller sensor is the watchdog that can monitor the environment and wake up the larger sensor when necessary."
While monitoring bridges is one application these remote sensors can be used for, other uses include predicting natural disasters or observing gases emitted by active volcanoes. These sensors require researchers to take long, arduous hikes to reach a location while wearing protective equipment to prevent damage to the skin and lungs as well as to protect against exposure to extreme heat and gases.
The device from the University of Pittsburgh could make these trips less frequent and may ultimately allow these devices to be powered by the environment.
"One of the main challenges of running AI algorithms with energy harvested from the environment is that the energy from the environment is intermittent," Hu said. "Much like a laptop, if the sensor loses power, you lose the data, so we want to help AI algorithms reach an accurate decision, even with intermittent power. By applying AI, we hope to increase the lifespan of unattended sensors and make them more reliable and useful.”
