Optimization of Traffic Signal Control Based on Game Theoretical Framework

Abstract

This paper presents a model of intelligent traffic signal control using a game theoretical framework based on decision-making operations. This model aims at finding an optimal solution to reduce traffic congestion and minimize the waiting time of cars in the urban transportation system. The expected strategies can be generated in the game theoretical framework to control the traffic signals for each incoming links (i.e. sections) in the intersection so that the vehicles are distributed optimally. In this system, each incoming link is regarded as a single player and the status of signal light (green or red) can be considered as the decisions made by these players. Thus, the global optimal solution for selecting the best routes of vehicles comes out by combining all the decisions and evaluating the cost function in this game. Furthermore, the effectiveness of the global optimal strategy and Nash equilibrium strategy produced from a game theoretical framework and constant strategy (i.e. the time interval of red and green lights is always constant and periodical) are compared after implemented in one intersection.