Design of Resilient Smart Highway Systems with Data-Driven Monitoring from Networked Cameras

This project aims to develop a systematic way to design smart highway systems with networked video monitoring and control resiliency against environment disruptions and sensor failures. On the video monitoring side, the research team will investigate 1) efficient deep learning methods for extracting fine-grained local categorical traffic information from individual surveillance videos (e.g., traffic mixture, environment information, anomaly/extreme-weather detection in the scene), and 2) novel graph neural network (GNN) methods to correlate and propagate the local information through the highway network for global states estimation (e.g., vehicle tracking and reidentification, traffic prediction in unobserved area). The team will discretize the traffic volume of a local road intersection into several levels/categories (e.g., low, mid, high), which will make the method less prone to estimation errors that could originate from challenging environmental variations. On the system design side, the team will 1) establish dynamic models for capacity using video data, 2) model failure in either cyber or physical components, 3) study the relation between sensor deployment and observability for resilient traffic control (e.g. route guidance and ramp metering). The expected outcome is an implementable approach to designing resilient smart highway systems with trustworthy monitoring capability. The team also expects their approach (with appropriate modification) to be applicable to general transportation systems.


  • English


Subject/Index Terms

Filing Info

  • Accession Number: 01705185
  • Record Type: Research project
  • Source Agency: Connected Cities for Smart Mobility towards Accessible and Resilient Transportation Center (C2SMART)
  • Contract Numbers: 69A3551747124
  • Files: UTC, RIP
  • Created Date: May 22 2019 1:12PM