A coordination model of game theory for multi-intersection-agents and the algorithm for solving equilibrium based on the reinforcement learning method

The complex transport facilities and conditions lead to the competition among different traffic intersections under the conditions of limited transportation resource. So the traffic intersection coordination is a game problem. Combining the game theory and reinforcement learning method, we propose a traffic intersection coordinating game model and solve the equilibrium of game by using the reinforcement learning method in this paper. The equilibrium means comprehensive balance optimum scheme of the whole coordination, which can optimize the traffic signal control of the objective area. Through the coordination of areas, balanced optimization of the whole urban transport system will be fulfilled. An experiment is presented to prove the algorithm is effective.