Machine Learning Applications in Vehicular Traffic Prediction and Congestion Control: A Systematic Review

In the context of large cities, the growth in traffic congestion has become a major problem. Static control systems may block emergency vehicles due to traffic congestion. The goal of detecting, forecasting, and reducing traffic congestion is to improve the level of service provided by the transportation system. One of the main issues that automatic way of traffic management plan will address is the current urban traffic congestion. Due to its capability to cope with dynamic behavior over time and with vast numbers of parameters in massive data, neural networks (NN) and machine-learning (ML) techniques are being employed more and more to tackle real-world issues, surpassing analytical and statistical methods. Therefore, deep learning is becoming more relevant for these jobs as access to bigger datasets at greater resolution increases. This study focuses on a critical assessment of the state of the art for the use of Artificial Intelligence in this specific field of Intelligent Transportation Systems. The literature that is now accessible uses a variety of methods to identify and dassify traffic congestion. The conclusions obtained from a review of recent articles using two taxonomic criteria are finally presented.