A reinforcement learning based traffic signal control algorithm in a connected vehicle environment

Abstract

The emerging vehicle technologies, i.e. connected vehicle technology and autonomous driving technology, can be beneficial for traffic control and operations. They not only serve as new source of information, but also enable us to control and modify the trajectory of vehicles. Reinforcement learning is widely used to design intelligent control algorithms in various disciplines. This paper provides preliminary results on how the reinforcement learning methods perform in a connected vehicle environment.