Dr. Zhe Fu | Postdoctoral Researcher, Stanford Energy

Time: 1:30 - 2:15PM

Date: Friday, October 2, 2026

Location: Hybrid, CE-CERT RM 105 & Zoom (Join Zoom Meeting ID: 92505532703; Passcode: 433385)

Title: Physics-Informed Learning and Control for Mixed-Autonomy Systems: Enabling Traffic Smoothing with a Few Automated Vehicles

Abstract: Mixed-autonomy systems, where automated and human agents coexist, are already emerging in real-world cyber-physical systems. A key challenge in these systems is how to leverage a small number of automated agents to influence overall system behavior under nonlinear dynamics, behavioral uncertainty, and partial observability. In this talk, I present a framework that integrates physics-informed learning, control design, and real-world experimentation for mixed-autonomy systems. Using traffic flow smoothing as a specific case, I show how Neural Finite Volume Methods enable accurate and data-efficient modeling of traffic dynamics, and how kernel-based and imitation-learning control strategies allow a few automated vehicles to dissipate stop-and-go waves while maintaining throughput. These methods are validated through simulations, small-scale closed-track experiments, and the largest scientific traffic field experiment to date, involving 100 automated vehicles on public highways, demonstrating measurable improvements in traffic stability and energy efficiency.

Bio: Zhe Fu is a Stanford Energy Fellow, hosted by Prof. Eric Darve and Prof. Marco Pavone. She received her Ph.D. in Transportation Engineering and M.S. in Electrical Engineering and Computer Sciences (EECS) from UC Berkeley, where she was advised by Prof. Alexandre Bayen and conducted research with the Berkeley Artificial Intelligence Research (BAIR) Lab.
Her research focuses on learning, modeling and control for distributed parameter systems, with an emphasis on mixed-autonomy systems. She develops physics-informed neural models for hyperbolic partial differential equations, designs model-based and data-driven control algorithms, and validates these methods through both closed-track and open-road large-scale field experiments. Zhe was named a 2025 Eno Fellow and the 2026 ITS Berkeley Outstanding Ph.D. Student of the Year, and she was the runner-up in the 2025 Berkeley Grad Slam. Her research has also received recognition across multiple communities, including first place in the 2023 INFORMS Poster Competition and Rising Star distinctions in Mechanical Engineering (Carnegie Mellon University, 2024), EECS (MIT, 2025), Cyber-Physical Systems (NSF, 2025) and Data Science (University of Chicago, 2026). Her contributions to leadership, mentorship, and teaching have been recognized by UC Berkeley and organizations including ITS, CTF, WTS, EDGE in Tech, H2H8, and AAa/e. Learn more at fu-zhe.com.