Enhancing Social Digital Twin (SDT) Models for roadways with satellite imagery data

PI:  Sean Qian

University:  Carnegie Mellon University

Industry Partner(s):  Fujitsu Research of AmeInc.

With the collaboration of a Fujitsu research team in Fujitsu’s Pittsburgh office and CMU’s Mobility Data Analytics Center, this project uses satellite imagery data to enhance the accuracy and granularity of travel demand data, including density metrics of driving vehicles, parked vehicles and non-motorized users. Fujitsu research has intensive knowledge and experience in processing satellite images and translating those to vehicular information. Those additional data will be integrated into MAC-POSTS, an open-source mesoscopic network backbone of a SDT platform developed by CMU, particularly enhancing model calibrations (e.g. O-D estimation, traffic flow fundamental diagrams, etc.). In addition, we will engage local stakeholders and CMU researchers to determine real-world use cases for SDT. This project will enhance the SDT model we collaboratively established (through PITA FY23) for Southwestern Pennsylvania, focusing on improving modeling accuracy and developing case studies for other global communities.