There’s still much work to be done before our buildings are smart enough to talk to us. But from a lab located in the most accelerometer-instrumented building in the world, Rodrigo Sarlo is doing his part to get us there.

Sarlo, a Ph.D. candidate in the Department of Mechanical Engineering at Virginia Tech, was recently awarded a research fellowship from the American Society for Nondestructive Testing. His work on the intelligent infrastructure project is now underway.

Sarlo’s research for the fellowship has far-reaching implications that could eventually help in designing buildings that are better able to find, analyze, and communicate information about the structure — a first step in making buildings communicate relevant information to their inhabitants.

Sarlo works in the Smart Infrastructure Lab, commonly called VT-SIL. The lab, founded and directed by Pablo Tarazaga, associate professor in mechanical engineering and John R. Jones Faculty Fellow, is located on the third floor of Goodwin Hall — and not by accident, since the building serves as their subject.

“We set out to make Goodwin Hall a very special and unique building. We made the effort to set up a system that would provide a truly multidisciplinary and adaptable test bed for research at the intersection of multiple fields,” Tarazaga said. “Today we have an operational real-world living laboratory where we conduct smart infrastructure research across multiple departments and multiple colleges, as we become a destination area in Intelligent Infrastructure for Human-Centered Communities.”

Outfitted with more than 200 accelerometers and other sensors, the 160,000-square-foot, five-story, L-shaped building collects 12 gigabytes of data each day: data that can be used to track the flow of occupant movement and their activity, to identify occupants based on gait, and to monitor structural health, to name a few applications.

It’s Sarlo’s job to help with the data acquisition, storage, and processing — particularly the latter two. Sarlo conducts full structural analyses on Goodwin Hall and works to keep the network of sensors and data acquisition units continuously monitoring.

Sarlo is also researching methods of using the data to better characterize the building’s vibrational response, which is influenced by a variety of occupational and environmental factors, like crowds inside the building or wind and earthquakes.

Estimating key structural parameters, like natural frequencies, provides information about the structural health of the building, but it’s no simple task, even in what is believed to be the most-instrumented building in the world. In fact, depending on the analysis being conducted, sometimes more sensors can complicate the process.

Through the fellowship, Rodrigo will work on designing an intelligent system that can pinpoint which sensors and which data should be collected and analyzed for the most-accurate estimate — that is, with higher confidence and narrower uncertainty — of the building’s structural information.

“I’m looking at: What are the challenges that buildings pose for producing accurate measurements and how can we address these challenges for various applications?” Sarlo said.

Establishing this kind of system would be applicable beyond Goodwin Hall. Not only could it help with damage detection in a building, having more-certain structural data could help engineers validate virtual models of a building, for example.

“A lot of these decisions, when they build the building are done based on some virtual model,” Sarlo said. “Very few times do they actually go back and assess the actual building and how it’s behaving.”

This possibility in particular could have massive implications for how civil engineers design buildings. Rather than using a virtual model that meets established codes, designing with the kind of data Sarlo’s research might uncover could help engineers make more-precise design decisions.

In studying and better understanding the way buildings behave in real-time, future buildings — and infrastructure in general — could be safer than standards of code anticipate.

“In fact, the civil engineering community is moving in that direction,” Sarlo said. “They’re trying to get away from code-based design, where you have a very specific set of criteria that you need to meet, and they’re moving into what is called performance-based design, which is a little more abstract.”

The possibilities of a system like the one Sarlo is working on pave the way toward futuristic, intelligent infrastructure. If an instrumented building knows which data to collect and when and can then act on that information, buildings can operate more efficiently.

“In a sense, you could think of the building as a robot; it’s able to process its own data, think for itself, and make decisions. At least that’s what we’re striving for,” Sarlo said.

Take the example of a fire inside a building. An intelligent building would know which data to collect to identify changes in the building’s structure.

Based on that change, the building could locate the fire. Simultaneously, the smart building would be able to call for first responders while also locating occupants — ideally down to the individual level. The building could then inform occupants of their path to safety. For those who have no safe path out, first responders would be able to locate them wherever they are in the building.

It’s a distant reality, but still in the realm of possibility for Sarlo, whose work with the fellowship and past research work continue to move the needle forward.

“That’s kind of the futuristic vision for this building. Obviously, there’s still a lot of work to do,” Sarlo said.

Written by Erica Corder