Implications of Information Structure in Control of Urban Traffic Networks

Rapid advancements in technology have facilitated a tremendous increase in the number of control/decision and sensor points in urban trac networks, ranging from an individual driver carrying smart phone to ramp meters to city-scale trac control center. Due to the large volume of data generation, it is computationally, and arguably even technologically, infeasible to inter-connect all the points to each other for real-time applications. Therefore, it is of interest to study performance of traffic networks under various information structures, i.e., sparse interconnection of control/decision and sensor points. In this project, we propose to study such issues under two complementary topics. First, we consider optimal control of traffic flow over networks using a combination of variable speed limit, ramp meter, lane-changing, and routing control. While this problem has attracted significant attention, most of the prior work has been limited to centralized or open-loop control. We propose to develop the foundations for a framework to design closed-loop control under given information structures. Our emphasis will be on computational tractability and characterization of performance gap with respect to centralized control. Second, we propose to study optimal information design to influence route choice decisions of drivers in dynamic environments. Specifically, we adopt the framework of algorithmic persuasion, under which the system planner can exploit information asymmetry about the knowledge of the real-time state of the network to release noisy information or recommend routes to the drivers in order to optimize social objective. The study of algorithmic persuasion in the context of routing games is very recent, and more so, the existing work implicitly assumes the drivers to evaluate the incentive compliant nature of the recommendations from the system planner only asymptotically, they do not consider externality from drivers who do not participate in persuasion, and assume static traffic flow models. In this project, we propose to address these shortcomings to develop foundations for algorithmic persuasion in routing games. The methodological contributions will be supplemented with case studies using traffic data from the Los Angeles area, and with simulation case studies in VISSIM.


    • English


    • Status: Active
    • Funding: $99,999
    • Sponsor Organizations:

      California Department of Transportation

      1227 O Street
      Sacramento, CA  United States  95843
    • Managing Organizations:

      METRANS Transportation Center

      University of Southern California
      Los Angeles, CA  United States  90089-0626
    • Project Managers:

      Brinkerhoff, Cort

    • Performing Organizations:

      University of Southern California, Los Angeles

      University Park Campus
      Los Angeles, CA  United States  90089
    • Principal Investigators:

      Savla, Ketan

    • Start Date: 20190801
    • Expected Completion Date: 20200731
    • Actual Completion Date: 0
    • USDOT Program: University Transportation Centers
    • Source Data:

    Subject/Index Terms

    Filing Info

    • Accession Number: 01708671
    • Record Type: Research project
    • Source Agency: METRANS Transportation Center
    • Files: RiP
    • Created Date: Jun 25 2019 4:13PM