UIC College of Engineering researchers selected for Discovery Partners Institute seed funding

Four projects involving College of Engineering researchers received funding from the first round of UIC’s DPI Cycle 1 Seed Funding Program. Chancellor Michael D. Amiridis and Vice Chancellor Joanna Groden budgeted up to $100,000 per project to catalyze UIC-led activities that advance the mission of the University of Illinois System’s Discovery Partners Institute (DPI). Up to $10 million will be leveraged from UIC’s royalties’ revenue for projects aligned with DPI goals.

Established in 2018, the DPI statewide network will partner research universities and industry, governmental, non-governmental, community-based agencies, and cultural and philanthropic organizations to advance technological innovation and grow the economy.

“We are very excited by these seed funding projects that span a variety of domains such as healthcare, autonomous vehicles, and transportation infrastructure. These projects are excellent illustrations of how the faculty from UIC’s College of Engineering are advancing DPI’s mission with their deep research expertise,” said Venkat Venkatakrishnan, associate dean for research and graduate studies for the college.

Of 88 proposals submitted for consideration, 10 were selected based on a combination of exceptional intellectual merit and research excellence, and potential for commercial impact. Below is a list of the four funded projects involving UIC college of engineering researchers:

“A Contextualized Dialogue Agent to Address Health Disparities among Urban African American and Latinos with Heart Failure,” Barbara Di Eugenio, computer science, with collaborators from the college of medicine and the college of applied health sciences.

More than half of heart failure patients lack the knowledge to manage their disease, and health care providers often use jargon-filled language when communicating with patients. African Americans are at the highest risk of developing heart failure, followed by Hispanics. By teaming up with health care coaches who work with minority populations, Di Eugenio and her UIC colleagues are designing a dialog agent that is adept in suggesting sound health care strategies that will be available to health care providers 24/7.

“The tool will be like an Alexa or Siri, which is competent in specific health care issues, but can use the right terms and be appropriate for these populations,” said Di Eugenio. “We will work with health care coaches and focus groups to discuss the types of questions patients have about heart failure, medications and other issues.”

“Control of High-Speed Autonomous Vehicles in Complex Environments Using Onboard Computing,” Shuo Han and Milos Zefran, electrical and computer engineering.

Autonomous vehicles (AVs) must solve complex optimization and decision problems in real time using a computer. These calculations must be made faster than the environment they are operating in changes, which is challenging especially when driving at a high speed on a highway. To tackle this challenge, Han and Zefran want to design a framework that is inspired by the way human drivers operate—reacting to the most pressing constraint, such as the nearest car, rather than trying to consider all factors, which is what most AVs do. At the same time, the system must ensure that safety is not compromised as the focus of attention shifts.

“This DPI grant will allow us to improve the reliability of control algorithms used in autonomous vehicles to ensure safe operation in a multitude of situations that may occur on the road. It will also allow us to create a research program that could potentially attract collaborations with industry through the new Discovery Partner Institute,” said Han.

“Ubiquitous Radar Systems for Autonomous Vehicles and Advanced Safety in Urban Environments,” Mojtaba Soltanalian and Amit Trivedi, electrical and computer engineering.

Radar-based safety technology is superior to visible and infrared imaging techniques for use in autonomous vehicles (AVs), and performs better in poor weather. The cost of ultra-high frequency radar signal processors prevents their mass use—the typical AV radar bandwidth is large, which makes it expensive. Soltanalian and Trivedi are using the machine learning framework of deep unfolding to achieve high performance, while using low-cost and low-speed technologies.

“The goal of our research is to reduce the cost of such radar systems while maintaining their reliability—thus helping with the mass deployment of radar-based advanced vehicular safety features. Given such requirements, our work will lead to novel hardware implementations that demonstrate our approach for on-field operations,” said Soltanalian.

“MY-AIR: Monitoring Your Air-Pollution Risk a Mobile App,” Jane Lin, civil and mechanical engineering, with collaborators from computer science and the college of medicine.

In the United States, 166 million people live in areas with unhealthy air, and the World Health Organization identified air pollution as the world’s largest single environmental health risk. While the U.S. Environmental Protection Agency’s AirNow program provides air quality information and associated health risk indicator – Air Quality Index (AQI) – to the public, more than 42 million people reside in places that are farther than 40 km from the nearest PM2.5 monitor.

To address this issue, Lin and her partners are developing the MY-AIR: Monitoring Your Air-Pollution Risk a Mobile app, which will reflect individuals’ health risk to pollution exposure and take into account a person’s microenvironment, activity, and physiology. It will consist of a smartphone app and enabling models and algorithms on the server end. MY-AIR will apply advanced deep learning techniques to interpolate current and predict near-future air pollutant levels at fine spatial and temporal scales, and provide personalized AQI.