Modeling Humans: Adaptive Control, Neural Networks, and Cyber-Physical Human Systems

Associate Professor Yıldıray Yıldız

Bilkent University

 

Date/Time: Thursday, 15 May 2025, 16:30

Place: SC106 Aysel Sabuncu Brain Research Center Seminar Room

Understanding and modeling human behavior is key to building smarter, safer, and more intuitive Cyber-Physical Human Systems. In this talk, I will present adaptive control strategies for modeling human pilots and drivers, with a focus on capturing real-time decision-making and interaction with autonomous systems. I will also introduce neural network-based human-operator assistance techniques that predict and adapt to operator behavior, improving stability and responsiveness. Finally, I will discuss applications in shared autonomy, where the authority is optimally distributed between humans and machines to achieve a smooth operation. Results demonstrate significant improvements in safety and performance, pointing toward a future of harmonious human-machine cooperation.

 

Bio:

Yildiray Yildiz is an Associate Professor of Mechanical Engineering at Bilkent University, where he leads the Systems Laboratory. His research centers on modeling human-operator behavior and developing adaptive control architectures for Cyber-Physical Human Systems (CPHS). Through his work, he aims to bridge the gap between human decision-making and autonomous control, enhancing safety and performance in applications ranging from operator assistance to autonomous driving.
Dr. Yildiz received his Ph.D. in Mechanical Engineering from MIT and subsequently served as a researcher at NASA Ames Research Center. He currently serves on the editorial boards of IEEE Transactions on Control Systems Technology and the European Journal of Control. His commitment to advancing CPHS research was further highlighted by his role as the General Chair of the 2024 IFAC Cyber Physical Human Systems Conference.