Inventing the Interconnected Future

Inventing the Interconnected Future

Doctoral students Alena Rodionova and Matthew O’Kelly are developing the Formula 1/10 autonomous race car’s sensor fusion module, part of the driver’s license test for the Safe Autonomous Vehicles project.

Anesthesiologists carefully monitor an infant during surgery to assure she receives a steady flow of oxygen. Yet should anything go awry, by the time the pulse oximeter on her finger indicates a falling oxygen level, she may already be in danger.

To develop a more protective earlier warning system, engineers with Penn Research in Embedded Computing and Integrated Systems Engineering (PRECISE) worked with clinicians to detect and use diagnostic clues in multiple sources of streaming physiological operating room data. “We focus on the life- and safety-critical medical, transportation and energy sectors, the ‘cyberphysical systems’ comprised of sensors, computation, control and communication that are tightly integrated with the physical world,” says Insup Lee, director of PRECISE and Cecilia Fitler Moore Professor in Computer and Information Science.

Infant operating room data has been used to develop a prototype predictive device for hypoxemia, an abnormally low concentration of blood oxygen. “Clinicians will be able to use this device to find predictive patterns for when a surgical patient of any age is at risk for hypoxemia so they can intervene earlier to prevent problems,” says Allan Simpao, an attending anesthesiologist at Children’s Hospital of Philadelphia. “The PRECISE team is fantastic at what they do and at interfacing with clinicians.”

Real-world impact like this abounds at PRECISE, a hub of theoretical and applied research by nine Penn Engineering faculty members who work across disciplines to advance solutions to urgent challenges and opportunities as the world shifts to sensor-based technologies. They lead $4 million per year in funded research and have pioneered new course and degree offerings that attract and cultivate next-generation leaders for a more connected and sensor-infused future.


Sensor-driven phenomena such as the internet of things (IoT) are not about a specific technology. Instead, IoT refers to the sensor-embedded household, industrial, medical and commercial devices that are increasingly being designed to communicate and coordinate virtually.

“PRECISE is at the forefront of research to make devices connect and integrate seamlessly while always considering the risk aspect — to ensure safety, security, privacy and real-time performance,” says Lee. “For instance, we’re developing techniques to help self-driving cars withstand ‘spoofing’ attacks that send counterfeit signals to a vehicle’s navigation system.”

Doctoral students Alena Rodionova and Matthew O’Kelly are developing the Formula 1/10 autonomous race car’s sensor fusion module, part of the driver’s license test for the Safe Autonomous Vehicles project.

George Pappas, Joseph Moore Professor and chair of Electrical and Systems Engineering (ESE), describes sensor-based phenomena like the internet of things as the latest technology revolution. “IoT will have widespread academic, industrial and corporate impact because it’s so big, but nobody has access to the full stack of that vision yet. The real breakthroughs over the next decade will occur if someone creates an operating system that allows us to program across many heterogeneous devices,” he states.

“This is an exciting time to go into this field,” concurs Linh Thi Xuan Phan, assistant professor in Computer and Information Science, who works on methods to defend interconnected devices against attacks, bugs and intrusions, and ways to assure that interconnected devices like cars and cardiac monitors maintain life-saving responsiveness. “It’s inspiring to invent solutions to shape a more safe and secure future.”


As he reviews real footage of fatal autonomous car crashes and describes the absence of a comprehensive regulatory framework, it’s clear there’s a need for the research conducted by Rahul Mangharam, associate professor in ESE. “We use simulated hyper-realistic environments with high fidelity physics to test drive real autonomous vehicle software and investigate how the vehicle behaves in accident and near-accident situations,” he says.

Mangharam connects data from autonomous cars to an adapted version of the video game Grand Theft Auto to test how autonomous cars respond to high-risk driving scenarios and improve vehicle perception, planning and decision control. Mangharam’s research collaborations focus on the technological, legal and ethical design of safe autonomous systems (such as how to develop driver’s license tests and insurance coverage for autonomous vehicles), strategies to manage energy consumption during volatile energy markets and weather patterns (with National Science Foundation (NSF) funding for commercialization of this approach), and medical topics such as integrating operating room device data to convey the fusion of factors affecting patients.

David Corman, Cyber-Physical Systems program director in the Computer and Information Science and Engineering Directorate at the NSF, says, “We would bet our lives on many of the devices PRECISE has worked on. Some of the areas of research they’ve engaged in are really new methods to verify that criticality. The PRECISE center is among the leaders setting this research agenda.”


“PRECISE is also among the groups that helped shape novel and innovative educational curricula in the area of IoT and cyber-physical systems,” Corman continues. “The students from their program have gone on to become some of the great investigators in this field.”

The Center is a hub for 15 postdoctoral fellows, 45 doctoral candidates, 40 master’s students and a growing number of undergraduates. Integrative degrees reflect how the program has anticipated the skills needed for this emerging field. In 2009, Penn began offering a master’s in Embedded Systems, one of the world’s first IoT-related educational degrees. A new IoT concentration for undergraduates is being developed in ESE and will be available beginning in 2018–2019, and an online micro-master’s IoT degree (via is being developed and will likely launch in 2018. Pioneering IoT course offerings are growing as well, such as Embedded Software for Life-Critical Applications; Digital Twins: Model-based Embedded Systems; and The Security of the Internet of Things, which is the first on this topic at any university.

“While our curriculum reform is never complete, we are way ahead of the competition in educating 21st century engineers for IoT,” continues Pappas. “We’re preparing our students to be innovators and participants in the enormous startup opportunities in the IoT space.” Striking a note of cautious optimism, Mangharam adds, “We’re mentoring the coming generation of engineering citizens who will develop these large-scale autonomous systems. These will soon become integrated into our lives in ways that we cannot take lightly, and the burden is on us to make sure these systems are safe and have a positive impact.”