Real-time Queue Length Estimation from Video Surveillance
Project ImageDescription:
Data Extraction: Analyzes video footage to extract vehicle trajectories and queue patterns.
Model Integration: Combines trajectory data with following models to reconstruct queue states dynamically.
Challenges Addressed: Mitigates issues like occlusions in video feeds and dynamic variations in traffic flow.
Applications: Optimizes red light timing and reduces congestion at high-demand intersections by accurately estimating queue lengths.
Yunwen XU
Associate Professor
Yunwen Xu is an Associate Professor at the Department of Automation, Shanghai Jiao Tong University. She earned her Ph.D. and M.S. in Control Science and Engineering from Shanghai Jiao Tong University and her B.S. in Automation from Nanjing University of Science and Technology. She was a visiting scholar at Purdue University and has held postdoctoral and research positions at Shanghai Jiao Tong University. Her research focuses on predictive control applications, intelligent transportation systems, robot control, embedded predictive control, and intelligent control of complex systems. She has published extensively in top journals and conferences and has been involved in several significant research projects.