With everything going "smart," the increased reliance and interaction between humans and technology is not just inevitable; it is our new reality. How humans and buildings interact is the focus of a critical, emerging field called human-building interaction (HBI). This was also the topic of a two-day, NSF-funded workshop held from May 30 to 31. The workshop was organized by Stephen Schrank Early Career Chair and USC Viterbi Professor of Civil and Environmental Engineering Burcin Becerik-Gerber, USC Viterbi Research Assistant Professor of Computer Science and Civil and Environmental Engineering Gale Lucas and UC Merced Professor of Cognitive and Information Sciences Ramesh Balasubramaniam. The workshop was part of the NSF's M3X: Mind, Machine, Motor Nexus” program.
Experts gathered represented a broad range of fields including computer science, architecture, occupational science, psychology, civil and electrical engineering. Together, they examined questions critical to defining how to move forward in developing HBI. Their goals spanned improving work performance, engendering better health and well-being and promoting best practices that reduce adverse environmental impacts.
At the same time, USC Professors Burcin Becerik-Gerber and Gale Lucas launched the USC Center for Intelligent Environments (CENTIENTS). The center supports ground-breaking research and brings together the world’s foremost leaders, scholars, innovators and intellectuals to engage in dialogue and debate, as well as form partnerships that promote human-centered design and integration of intelligent technologies into built environments.
Listen to the podcast inspired by Burcin Becerik-Gerber’s work.
Speakers at the workshop included Allen Sayegh, associate professor in practice of architectural technology at Harvard University, and Shawn Roll, director of the Ph.D. in occupational science program at USC. Topics included: computer simulation of crowd behavior, privacy and security, spatial comfort, interactive architecture, adaptive human-building coupling, sensing and learning for HBI, navigation and spatial cognition and user psychology.
"Where should we aim between fully manual and fully automated? If you ask computer scientists, they say fully automated. If you ask architects, they say fully manual. I suspect the answer is somewhere in between," said Liam O'Brien, associate professor in architectural conservation and sustainability engineering at Carleton University, who spoke about understanding and designing for occupant behavior.
In researching user preferences, O'Brien found that uncomfortable users will always find ways to change controls in their favor.
Take, for instance, the photo on the right--when building managers cut occupants off from changing the thermostat to match their preferences, occupants found a tasty workaround: freezer pops.
"As we think about smart cities, what is the source of real challenge and complexity? We came to thinking about complexity inherited boundaries that accompany instances when you take on technology on a large scale. For instance-- security or organization."
--Bhaskar Krishnamachari, USC Viterbi professor of electric and computer engineering and computer science and Ming Hsieh Faculty Fellow in Electrical Engineering Systems discussed key issues related to centralizing smart data sources.
"The internet is currently built out on a 'narrow waist architecture,'" Krishnamachari said. "Everything runs on IP and IP runs on everything. Currently the internet of things is still application by application. But that’s not the way the internet was built and that’s not fundamentally the way we can scale things."
Looking to the future, Krishnamachari and his team are working on a similar idea for smart apps. Each will connect to the same middleware, centralizing and streamlining data sourcing and mediating issues like cybersecurity and privacy.
"Architects are confined to manual exploration of these spaces. This is very limiting. To that end, we take advantage of the fact that this part of the pipeline can be automated. We can use AI to explore the space and provide suggestions on how environments can be reconfigured."
- Mubbasir Kapadia, assistant professor in computer science at Rutgers University
In concluding the workshop, the group determined ways to begin opening communication channels to help cross-pollinate ideas, share research and fuel future collaborations. As challenging as it is to come together across different fields, the value of such collaboration was imminently clear for the future of smart buildings.
The possibilities, as they say, are endless to build dynamic spaces that fundamentally improve the way we live and work.
Slides provided courtesy Liam O'Brien, Mubbasir Kapadia and Simi Hoque. Additional photos by Avni Shah.