Date and time: April 16, 2026, 2:00 PM CDT
Speaker: Chuheng Wei — University of California, Riverside
Archive: Flier | Recording
Register today
About the talk
This seminar presents recent research on intelligent and personalized autonomous driving systems integrating robust perception, cooperative intelligence, and human-aware prediction in complex real-world environments. Autonomous vehicles must operate reliably under adverse weather, heterogeneous sensor configurations, and diverse human driving behaviors. To address these challenges, we develop end to end perception frameworks that improve object detection under non ideal visual conditions without separate image restoration, enabling stable environmental understanding in degraded scenarios. We further explore V2X cooperative perception through a unified heterogeneous fusion framework that supports diverse sensor configurations across vehicles and infrastructure, extending perception beyond line of sight while maintaining efficiency. Building on reliable perception and cooperation, the seminar introduces personalized trajectory and driver behavior prediction in complex traffic scenarios. Using a controlled multimodal driving dataset integrating vehicle dynamics, environmental sensing, and physiological signals, we study driver specific decision making and advance human-aware prediction for safer and more adaptive autonomous driving systems.
Speaker
Chuheng Wei is a Ph.D. candidate in Electrical and Computer Engineering at the University of California, Riverside. His research focuses on V2X-enabled cooperative driving systems, with emphasis on cooperative perception under non-ideal sensing conditions and data-driven modeling for multi-agent interaction, trajectory prediction, and connected and personalized automated driving. He has authored over 20 peer-reviewed publications in leading venues across intelligent transportation systems and computer vision, including papers in Proceedings of the IEEE, one of the oldest and highest-impact journals in the IEEE portfolio. He is also a co-author of a V2X cooperative perception method that won First Place in the CVPR 2025 challenge and received a Best Paper Nomination at IEEE ITSC 2025. He actively serves as a reviewer for major journals and conferences in ITS, robotics, and computer vision.