Urban Traffic Modeling based on Urban Pattern Clustering

In order to improve the prediction of MPC, we use urban patterns to clustering urban traffic daily volume data, and provide more information to predictive models. These predictive models are macroscopic models for control. There are several research interests for this proposal:

  • mixture model based urban traffic data clustering;
  • urban traffic network partitioning based on clusters;
  • fault detection of urban traffic;

Besides, we are using SUMO-SIM as our simulation platform, so a program that connects matlab/simulink and SUMO will be developed as well. This program will be used for many other further research.

Project members:
Y. Hu, MSc (Yu)

Model predictive control, Distributed and large-scale systems, Multi-level and multi-agent control