Agent-Based Game Theory Modeling for Driverless Vehicles at Intersections

Transportation engineers attempt to design highway networks to reduce the number of crashes and delay at the same time. Many artificial intelligence labs have suggested the use of fully driverless (autonomous) vehicles that have the capability of sensing the surrounding environment to enhance the roadway safety. A driverless vehicle that drivers in the traffic stream will have to do everything from obeying the speed limit, staying in its lane, detecting pedestrians and choosing the best route. Consequently, the automated vehicles should be provided with video cameras, radar sensors and a laser range finder and detailed maps to navigate the road ahead. The idea of having fully automated vehicles running in streets was inapplicable for many years till recently some of the research centers succeeded to release fully automated vehicles without human drivers (e.g. Google Driverless cars, October 2010). It is anticipated in the future that most of the vehicles will be fully automated; thus the movements of those vehicles will need to be optimized in the network. This research attempts to focus on modeling driverless vehicles at intersections as autonomous agents. The vehicles movements inside the intersections will be optimized using game theory algorithms in order to minimize delay and/or fuel consumption levels. There are a growing number of agent-based applications in a variety of fields and many transportation systems, including decision support systems, dynamic routing, congestion management, and intelligent traffic control. The use of agents of many different kinds in a variety of fields of computer science and artificial intelligence is increasing rapidly due to their wide applicability. Agent-based modeling "ABM" (or multi-agent modeling) has emerged as a modeling algorithm for modeling complex systems composed of interacting and autonomous units (i.e. driverless vehicles). Agents have behaviors, often described by simple rules, and interact with other agents, which in turn influence their behaviors. The level of an agent's intelligence could vary from having pre-determined roles and responsibilities to a learning entity. Consequently, agent-based modeling will help to maximize the efficiency of moving vehicles through intersections. It is assumed that intersections could be equipped with a dedicated wireless communication system and a protocol for communicating and giving permission to vehicles to pass. In the proposed system, vehicles must traverse intersections according to a set of parameters agreed upon by the vehicle and the intersection controller (decision maker). The vehicle movements through the intersection will be constrained by the vehicles' physical characteristics and weather conditions. The decision of passing or not for each vehicle will be given by the manager (controller) of the intersection based on a game theory algorithm. A game is simply defined as a conflict in interest among n individuals or groups (players). There exists a set of rules that define the terms of exchange of information, the conditions under which the game begins, and the possible legal exchanges in particular conditions. Game theory has been applied in many engineering, economics and biological fields. However, the decision making for driverless vehicle movements at intersections using game theory is considered an innovative idea in the transportation engineering arena. The game theory algorithm of this research is presented as a multi-objective function seeking delay minimization and/ or reducing fuel consumption. The decision maker -intersection manager- is modeled as a player trying to maximize the total benefit (delay and/or fuel consumption) for the entire intersection by giving the appropriate decision (passing or not) to each vehicle. In summary, the research attempts to optimize the movements of the future intelligent (driverless/autonomous/unmanned) vehicles at intersections as agents that have certain goals and limitations. Vehicle movements will be controlled using game theory algorithms in order to reduce the total delay and fuel consumption at intersections.


  • English


  • Status: Completed
  • Funding: $60000.00
  • Contract Numbers:


  • Sponsor Organizations:

    Mid-Atlantic Universities Transportation Center

    Pennsylvania State University
    201 Transportation Research Building
    University Park, PA  United States  16802-4710
  • Performing Organizations:

    Virginia Polytechnic Institute and State University, Blacksburg

    Virginia Tech Transportation Institute
    3500 Transportation Research Plaza
    Blacksburg, VA  United States  24061
  • Principal Investigators:

    Rakha, Hesham

  • Start Date: 20110101
  • Expected Completion Date: 0
  • Actual Completion Date: 20120830
  • Source Data: RiP Project 28380

Subject/Index Terms

Filing Info

  • Accession Number: 01474837
  • Record Type: Research project
  • Source Agency: Mid-Atlantic Universities Transportation Center
  • Contract Numbers: DTRT07-G-0003
  • Files: UTC, RIP
  • Created Date: Mar 7 2013 1:00AM