REU Mentor: Dr. Jinwoo Jang
This project advances the capabilities of data-driven digital twins to better understand infrastructure operations and human behavior within urban streetscapes. By feeding real-world sensor data into digital twin models, we enable highly realistic simulations and predictive analytics. In return, these digital replicas provide actionable insights into smart cities to enhance public safety, build resilient infrastructure, and optimize streetscape operations. The project involves the AI-centered development of: (i) a LiDAR sensing system to capture high-fidelity urban data, (ii) realistic traffic and pedestrian simulations, (iii) the integration of sensor data into digital environments, and (iv) high-fidelity, physics-based modeling. The intellectual merit of this research lies in the fusion of heterogeneous data (e.g., point clouds, video, trajectories) to achieve hyper-local streetscape understanding; the dynamic modeling of human behaviors and context interactions; and the realization of a photorealistic streetscape digital twin. The broader impacts are significant, as the outcomes will empower researchers, city managers, and elected officials to create safer and more enjoyable city environments for all.
Under the leadership of Dr. Jang, the project will provide a research experience for up to two REU participants. The first participant will focus on collecting and analyzing heterogeneous sensor data for streetscape spatial information. The second participant will focus on developing a photorealistic city model based on 3D map data, integrating coupled agents to enable streetscape simulation and forecasting.