A deterministic predictive traffic signal control model in intelligent transportation and geoinformation systems

Abstract

In this paper, we propose a traffic signal control method in intelligent transportation and geoinformation systems, based on a deterministic predictive model. The method provides adaptive control based on traffic data, including data from connected and autonomous vehicles. The proposed method is compared with the state-of-the-art traffic signal control solutions: empirical control algorithms and reinforcement learning-based control methods. An advantage of the proposed method is shown and directions of further research are outlined.

Keywords

• image segmentation
• road pavement distress
• synthetic dataset
• generative adversarial network
• convolutional neural network