CS3 Monthly Research Exchange

At each CS3 Monthly Research Exchange, three faculty and student researchers in the field of smart urban planning will take the stage to share their latest findings, breakthroughs, and urban projects.

Chinwendu Enyioha, Assistant Professor at University of Central Florida (UCF)

• Presentation Abstract: In the talk, I will present some results on fully distributed cooperative state estimation in communication-constrained environments. I will show how agents can effectively manage communication bandwidth by compress state information being shared with neighbors and summarize stability guarantees.

Zhaobin Mo, Ph.D. Student at Columbia University

• Presentation Abstract: For its robust predictive power (compared to pure physics-based models) and sample-efficient training (compared to pure deep learning models), physics-informed deep learning (PIDL), a paradigm hybridizing physics-based models and deep neural networks (DNN), has been booming in science and engineering fields. One key challenge of applying PIDL to various domains and problems lies in the design of a computational graph that integrates physics and DNNs. In other words, how physics is encoded into DNNs and how the physics and DL components are represented. We present a variety of architecture designs of PIDL computational graphs and how these structures are customized to different tasks in transportation, i.e., traffic state estimation and car-following behavior modeling.

Sonia Moshfeghi, Ph.D. Student at Florida Atlantic University

• Presentation Abstract: As in-vehicle telematics technologies advance, they offer applicable techniques for monitoring and evaluating driving behavior. Telematics systems can combine telecommunication and information processing to capture various types of driving data, such as acceleration, GPS, and onboard diagnostics (OBD) data. Identifying significant patterns and behaviors with feature extraction techniques provides insights into driver performance for applications like autonomous vehicles, demand estimations, insurance pricing strategies, driver training programs, and accident prevention strategies.

CS3 VALIDATE Accelerator Program:
Attendees of this month’s CS3 Research Exchange are welcome to stay after the presentations for an informational session (1:00 – 1:30PM) on the CS3 VALIDATE Accelerator Program.

The NSF Engineering Research Center for Smart Streetscapes (CS3) VALIDATE Accelerator is a 9-week non-dilutive program focused on validating and scaling successful early-stage ventures through in-depth education, mentorship, and business model design. It offers early stage teams a customized program with education from industry and academic research experts, hands-on value-added workshops, active mentoring, and venture creation opportunities.

The program will run February through April of 2024.

The CS3 Monthly Research Exchanges are internal and open only to CS3 affiliated students, faculty, and staff. If you are interested in learning more about the research happening at CS3, please contact our team at streetscapes@columbia.edu.