Real-Time Traffic Signal Control for Roundabouts by Using a PSO-Based Fuzzy Controller

Abstract

Developing traffic signal control methods is considered as the most important way to improve the traffic efficiency of modern roundabouts. This paper applies a traffic signal controller with two fuzzy layers for signalizing roundabouts. The outer layer of the controller computes urgency degrees of all the phase subsets and then activates the most urgent subset. This mechanism helps to instantly respond to the current traffic condition of the roundabout so as to improve real-timeness. The inner layer computes extension time of the current phase and decides whether to turn to the next phase in the running phase subset. As the phase sequences are well-designed, the inner layer smoothes the traffic flows which helps to avoid traffic jam. An offline particle swarm optimization (PSO) algorithm is developed to optimize the membership functions adopted in the proposed controller. In this way, the membership functions in the controller are no longer given by human experience, but provided by the intelligent algorithm. Simulation results demonstrate that the proposed controller outperforms previous traffic signal controllers in terms of improving traffic efficiency of modern roundabouts.