Real-time traffic signal control using deep Q network in V2X environment

In urban networks which mainly controlled by the traditional fixed signal control system, it would be impossible to manage real-time traffic situations properly. Furthermore, the autonomous vehicles would be impossible to exhibit its own maximum performance in this traditionally controlled intersection. In this point, more efficient traffic management could be possible through the AI traffic signal controller based on the real-time information. In this paper, on the premise of V2X technology, the Python algorithm was developed which controls the traffic signal in real-time by learning the traffic flow information of a single intersection. In the reinforcement learning, a particular signal light is displayed based on the information of each flow movement iteratively. The reward corresponding to changes in the network situation is given, and the learning proceeds. Through this study, it is expected to be foundation for achieving the system optimal in complex urban transportation networks.