A US-India team will jointly focus on a key application in smart cities – transportation management and will use the COSMOS testbed for experimentation. The team consists of US participants (PI: Prof. Sharon Di, co-PIs: Professors Qiang Du, Zoran Kostic, and Gil Zussman (Columbia)) and Indian participants (PI: Prof. Dhish Kuman Saxena, Prof. Amit Agarwal (IIT Roorkee), and Prof. Ashish Ghosh (ISI Kolkata)). The groups received supplemental funding from the US NSF and the Indian Technology Innovation Hubs (TIHs) supported by the Indian Department of Science and Technology under the National Mission on Interdisciplinary Cyber-Physical Systems, as outlined in the NSF US-India Collaborative Research Initiative. The 1-year effort titled “Networked Adaptive Traffic Signal Control in IoT-Enabled Smart Cities” is supported in the US by a $100K supplement to the ongoing project “NSF CPS: Medium: Hybrid Twins for Urban Transportation: From Intersections to Citywide Management”. The Indian team is sponsored under the TIH on Data Science, Big Data Analytics and Data Curation (Indian Statistical Institute, Kolkata).
The team will build on an ongoing NSF CPS project and will focus on transportation management. The goal is to develop a networked traffic signal control system enabled by Internet-of-things (IoT) in smart cities, leveraging two existing testbeds, the COSMOS testbed in NYC, and T-CPS testbed in Mohali, Punjab. The synergy between the US team and the India team lies at the intersection of data science and smart cities, including Computer Vision, Machine Learning, Data-Driven Optimization, and Digital Twin.
In particular, this project will leverage COSMOS to develop a digital twin for urban traffic systems, and use transfer learning to generalize the developed digital twin to simulate and optimize Indian traffic systems, calibrated with traffic data provided by the Indian testbed.