NYU Tandon research to improve emergency responses in urban areas with support from NVIDIA

A team of researchers from NYU Tandon’s C2SMARTER — a U.S. Department of Transportation-funded Tier 1 University Transportation Center — has received an NVIDIA Academic Grant Program award to develop an advanced simulation system aimed at improving emergency response in urban areas.

The project, called NanoDT (Nano Digital Twin), will create a detailed virtual replica (or “digital twin”) of Downtown Brooklyn that can help emergency responders better navigate accident scenes and manage traffic disruptions. The project is led by Principal Investigator (PI) Kaan Ozbay, Director of C2SMARTER and professor at NYU Tandon's Civil and Urban Engineering Department, along with Co-PI Zilin Bian, Research Scientist at C2SMARTER.

By integrating 3D visualization with real-time traffic data, the NanoDT system will offer a comprehensive and dynamic view of emergency situations to enhance situational awareness and response strategies.

While similar to other urban modeling projects, NanoDT specifically focuses on coordinating between emergency response and traffic management centers. The system aims to help cities better handle major disruptions, like the recent Baltimore bridge collapse or Philadelphia's I-95 highway collapse, which can cause severe casualties and economic losses.

"Emergency responders often rely on limited information from human reporting or fixed cameras," said Dr. Bian. "Our system will give them a bird's-eye view of accident scenes, helping them identify optimal routes and predict secondary accident risks."

"In urban emergencies, where every second counts, NanoDT's real-time 3D visualization will transform traditional response methods into dynamic, data-driven decisions," added Dr. Ozbay. "Ultimately, we want to help save more lives in critical situations."

The technology leverages existing city infrastructure, including traffic cameras and open data feeds, eliminating the need for additional investment in data collection. The system processes this existing data using artificial intelligence to create real-time 3D simulations of accident scenes, allowing responders to assess situations before arriving.

A key innovation is the system's ability to identify nearby resources like parking lots and urgent care facilities. "If there's a parking lot behind an accident scene, responders can park emergency vehicles there instead of blocking traffic lanes," Dr. Bian explained. This approach could help reduce secondary disruptions that often occur during emergency response.

The one-year project, starting this month, will tap into NVIDIA technologies including the NVIDIA RTX 6000 Ada Generation GPU, the NVIDIA Modulus framework, and the NVIDIA Omniverse platform. Modulus will be used for physics-based modeling and Omniverse will help with building 3D simulation workflows.

The project is currently in a proof-of-concept stage, with researchers — including Dr. Fan Zuo, a research scientist in C2SMARTER — using publicly available traffic and transportation data about Downtown Brooklyn to demonstrate its potential. If successful, the technology could be implemented in cities that already have extensive camera infrastructure, offering a cost-effective way to enhance emergency response capabilities.

“We are excited that NYU Tandon’s C2SMARTER is piloting NanoDT in Downtown Brooklyn,” said Downtown Brooklyn Partnership President Regina Myer. “Leveraging AI technology to analyze public data and better coordinate responses to major disruptions and emergencies is truly groundbreaking, and with its density of public transit and proximity to interstate highways and the East River bridges, Downtown Brooklyn is the perfect study area. This is a prime example of using New York City as a living laboratory to create solutions that make life better for all New Yorkers, and we look forward to the outcomes.”

NanoDT adds to C2SMARTER’s growing digital twin portfolio implemented in other parts of New York City, including Harlem — in a project with the New York City Fire Department to reduce emergency vehicle response times – and MetLife Stadium, in a project to help traffic management decisions both on and offline. 

“The C2SMARTER team will continue to advance transportation systems and engineering fields by using computationally demanding Artificial Intelligence/Machine Learning-based data acquisition and prediction approaches for congestion reduction and management in megacities such as New York City,” said Dr. Ozbay. “Industry collaboration with leading companies like NVIDIA is expected to play a key role in ensuring the success of these pioneering projects that will eventually save lives and money throughout the country.”

The NVIDIA Academic Grant Program advances academic research by providing world-class computing access and resources to researchers. Applications are now open for full-time faculty members at accredited academic institutions who are using NVIDIA technologies to process large- scale datasets, train graph neural networks, and accelerate projects in data analytics, statistical methods, robotics, automated vehicles, 6G, federated learning, and smart spaces.