Senem Velipasalar, Ph.D., Associate professor of electrical engineering and computer science, College of Engineering and Computer Science, Syracuse University.

“This kind of technology could have many other energy-saving applications.”

Project: Developing a low-cost, high-accuracy sensor platform that detects human presence inside buildings to reduce energy use in residential settings by as much as 30 percent.

Backstory: About 13 percent of all energy produced in the United States is used to heat, cool and ventilate buildings. Much of this energy is wasted by heating, cooling and over-ventilating unoccupied or partially occupied spaces. Existing building automation systems rely mostly on motion detectors and are limited in their reliability and ultimate ability to substantially reduce HVAC energy use.

Nuts and Bolts: Through a $1.2 million ARPA-E grant, Velipasalar is leading a team that partners faculty from SU’s Department of Electrical Engineering and Computer Science with SRI International, a leading nonprofit research center with expertise in embedded vision and machine learning. Their goal is developing a sensor platform using an infrared sensor, a visible-range camera, microphone and low-power processor to detect human presence— including in low light conditions and when people are static—and to develop algorithms to analyze and combine data from these sensors to enable occupancy sensing that would be impossible by each sensor alone. “We are making use of off-the-shelf components to develop a battery-operated, stand-alone platform that can perform occupancy detection in an efficient and reliable way on site,” she says.

SyracuseCoE Impact: Velipasalar, who has a strong record
of securing NSF support, credits former SyracuseCoE executive director Ed Bogucz with informing and motivating the team about this funding opportunity, as well as guiding them through the ARPA-E proposal preparation. “The ARPA-E proposal and budget preparation is different from NSF,” she says. “SyracuseCoE was very helpful every step of the way, especially in helping us develop and manage the budget.”

Practical Application: While the ARPA-E project is designed specifically to reduce HVAC energy consumption, Velipasalar says, “This kind of technology could have many other energy-saving applications,” including lighting and sound systems.

Intellectual Collision: Velipasalar’s research is at the intersection of embedded smart cameras, computer vision and machine learning. Her focus on questions related to energy is a more recent development, an outgrowth of her connection with SyracuseCoE. In addition to the ARPA-E project, an ongoing project with Tarek Rakha, former SU assistant professor of architecture and SyracuseCoE Faculty Fellow, investigates heat loss in buildings using thermal cameras on drones. Velipasalar and her Ph.D. student have developed an algorithm that autonomously detects heat leakages from thermal images of building structures. 

Led by Syracuse University, SyracuseCoE engages students, faculty, and collaborators at 200+ firms and institutions to catalyze innovations that improve energy efficiency, environmental quality, and resilience in healthy buildings and cleaner, greener communities.