Session: 20-17-01: Rising Stars of Mechanical Engineering

Paper Number: 172235

Data-Driven Multiscale Modeling of Complex Traffic Systems Utilizing Networked Driving Simulators 

This Faculty Early Career Development (CAREER) award supports foundational research focused on human driving behavior and the complex interactions among traffic participants, with the goal of advancing autonomous and connected vehicle technologies. A comprehensive understanding of driving behavior, both at the individual level and in interactions with other drivers, is essential for developing connected and autonomous vehicles (CAVs) that can safely and efficiently coexist with human drivers on the road. This project aims to deepen our understanding of how human drivers engage with their environment and with other road users, including the physiological and cognitive aspects of these interactions.

As the first part of the project, we have developed an immersive virtual reality (VR)-supported, multi-participant networked driving simulator setup to enable safe, scalable, and cost-effective experimentation. This system integrates motion platforms to deliver realistic motion cues, providing drivers with a more naturalistic driving experience. The simulation software enables the creation of diverse and customizable traffic scenarios, including complex and high-risk situations that would be unsafe or impractical to replicate on real roads. Multiple participants can simultaneously operate within the same virtual environment, where they can see and interact with each other in real time, allowing for the study of human–human interactions in traffic settings. This setup serves as a powerful testbed for controlled, repeatable experiments designed to investigate how driving behavior changes under different traffic conditions and in the presence of other drivers, while enabling data collection at the driver, vehicle, and brain (EEG) levels. To date, we have successfully completed the development of the system and achieved synchronization of multilevel data streams, including vehicle telemetry, driver inputs, and EEG signals—both within individual participants and across multiple interacting drivers. To support this integration, we developed a custom synchronization software, which we have made openly available to the research community.

A central focus of this CAREER project is the study of multiscale traffic interactions—spanning the vehicle level, the individual driver level, and the cognitive or brain activity level. The project aims to investigate how these layers of interaction influence one another and contribute to emergent traffic behavior. So far, we have validated data-driven approaches that can determine interactions from data, quantify the level of influence, and subsequently estimate and predict traffic behavior for individual driver-vehicle units. The next step of this project will use synchronized brain scans (e.g., EEG data) from multiple interacting drivers to study how cognitive engagement and mental workload correlate with observable driving behaviors such as reaction time, lane keeping, and interaction with other vehicles.

This research aims to uncover cognitive and behavioral markers that influence traffic dynamics, offering transformative insights for behavioral neuroscience and guiding the design of next-generation driver assistance and human-centered autonomous vehicle systems. Additionally, the project will produce extensive multiscale datasets, uncover modeling parameters grounded in experimental data, and lead to the development of experimentally validated traffic behavior models. Beyond research, the project integrates a strong educational and outreach component. It seeks to inspire student engagement in STEM disciplines and to conduct community-focused outreach to raise awareness about critical traffic safety issues, such as secondary crashes and impaired driving.

Presenting Author: Subhradeep Roy Embry-Riddle Aeronautical University - Daytona Beach

Presenting Author Biography: Dr. Subhradeep Roy received his Bachelor's degree in Mechanical Engineering from the Indian Institute of Engineering Science and Technology, Shibpur, in 2010, followed by a Master's from the Indian Institute of Technology Kanpur in 2012. He completed his Ph.D. in Engineering Mechanics at Virginia Tech in 2017. From 2017 to 2019, he worked as a post-doctoral scholar at the Physical Computing Laboratory at Virginia Tech, and between 2019 and 2021, he was a tenure-track assistant professor in the Mechanical Engineering Department at California State University, Northridge. In Fall 2021, he joined the Mechanical Engineering Department at Embry-Riddle Aeronautical University (ERAU), Daytona Beach campus, where he continues to serve as a tenure-track assistant professor. At ERAU, Dr. Roy directs the Complex Dynamical Systems Laboratory, where his research focuses on understanding complex dynamics in both human-made and natural systems. His work blends interdisciplinary and data-driven approaches to study how local interactions shape overall system behavior, with applications in biologically inspired swarms, brain connectivity networks, and traffic and transportation systems. Dr. Roy is the recipient of the 2023 NSF CAREER Award.

Authors:

Subhradeep Roy Embry-Riddle Aeronautical University - Daytona Beach