En Route.

towards the future of vehicular mobility!


Vehicular mobility is of utmost importance in our modern age. With rise of new TNCs (Transportation Network Companies) and advances of communication technologies, including smart and connected vehicles and cities, the shape of transportation is changing. We share our findings and data here!


We have just started and there are follow ups and journal versions yet to come, so for now, please have a look at our initial paper (please cite us if you use the paper, codes or any of the scenarios):

  • R. Ketabi, B. Alipour and A. Helmy, "En route: Towards vehicular mobility scenario generation at scale," 2017 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Atlanta, GA, 2017
  • Generated Scenarios

    Below is a sample of scenarios we generated. If you need other cases please contact us. I will add other cities and times of day over time! Feel free to contact me if you need help running them!

    • London (I think it was 10 to 10:30 morning!) scenario with lognormal parameters of u6.6, std0.6, 5shortest paths adn 2.5 for bmult using 10 iterations of DUA toward Equilibrium state: London_100cam_EQ10th.zip
    • Washington DC. Scenario (8 to 9 in the morning) with 100cameras and lognormal parameters of u7.0, std0.5 , 3shortet paths and no scaling for bmult (=1), using 10 iterations as well: WDC_8_9_EQ10th.zip


    Check our github page: github.com/rzbhk/EnRoute

    Sample data

    These sample data includes the densities extracted from the cameras using background subtraction (which I have used to generate the scenarios from). In other word, they have been the input to our En Route OD estimation sub system. It typically has camera definition and pairwise distances (in time, using maps) + densities over time. For OD and trip files check the scenarios.