Skip Navigation
Search

Bridging the Information Gap

Advancing Autonomous Systems Through Interactive Control

Professor Romeil SandhuThe College of Engineering and Applied Sciences announces that Professor Romeil Sandhu has been awarded $450K from the Air Force Office of Scientific Research (AFOSR), through the Air Force’s Young Investigator Research Program.  Professor Sandhu is an assistant professor in the Department of Biomedical Informatics, jointly administered by Stony Brook’s College of Engineering and Applied Sciences and School of Medicine.

AFOSR received more than 285 proposals for this prestigious early career award, open to scientists and engineers at research institutions across the United States who received PhD or equivalent degrees in the last five years. The program’s objective is to foster creative basic research in science and engineering, enhance early career development of outstanding young investigators, and increase opportunities for the young investigators to recognize the Air Force mission and the related challenges in science and engineering.

Professor Sandhu is one of only 43 scientists and engineers to receive the award, for his proposal addressing 3D Interactive Feedback Control for Autonomous 2D Imaging Systems.

“Rome is a brilliant young faculty with tremendous mathematical talent,” said Joel Saltz, MD, Chair of the Department of Biomedical Informatics. “I am extremely excited about this well-deserved award which will allow him to continue to generate groundbreaking new results.”

Sandhu’s research is focused on developing models and theory for understanding dynamical systems through the confluence of discrete geometry, statistics, and control spanning topics ranging from network science, computational vision, systems biology, and machine learning.  In particular, the AFSOR awarded Sandhu’s work in combating risk complexities we face in the real world when deploying autonomous systems such as unmanned (driverless) vehicles.  While there have been dramatic technological advances in machine learning and control approaches aimed at developing fully autonomous systems, user-experts implicitly “oversee” the automation process in the event of unintended events.

“Today, there is no universal machine learning algorithm capable of handling all real-world scenarios, so we often require humans to ‘interact’ with these systems to deal with the ‘unknown unknown.’  Whether it is protecting the soldier in the field, performing situational awareness, or self-driving cars, we have not reached a point in which we can trust complete autonomy over basic human input,” said Sandhu, who is also affiliated with the Departments of Applied Mathematics and Statistics and Computer Science. “Moreover, when we do apply control to drive these systems, we often derive our actions from observed image dynamics and tacitly ignore its 3D real-world counterpart.  Through this research, we are developing new theory from feedback control to allow for proper 3D human interaction from observed 2D sensor information”

“Artificial Intelligence is a strategic research thrust for our college.  Rome’s research will lead to breakthroughs that will enable humans to better interface with intelligent systems and will impact a wide range of AI-driven technologies,” said Fotis Sotiropoulos, Dean of the College of Engineering and Applied Sciences. “This prestigious award is a major career accomplishment for Rome  and I am really proud to have faculty of his caliber in our college.”

Romeil Sandhu Receives Young Investigator Award from
Air Force Office of Scientific Research