Traffic congestion frequently occurs in highly populated cities and can result from poor civil planning or inadequate public transportation. This issue increases traffic accidents, air pollution, fuel loss, and public dissatisfaction. Therefore, implementing traffic control systems that improve traffic flow and reduce travel times becomes essential. This work conducts a systematic literature review to identify the most efficient methods, algorithms, and models for developing traffic control systems. The review identifies three methods and three algorithms that are highly efficient for these systems, highlighting Bayesian filters and convolutional neural networks. It also shows that You Only Look Once (YOLO) is the most efficient image processing model for these implementations.

Traffic congestion frequently occurs in highly populated cities and can result from poor civil planning or inadequate public transportation. This issue increases traffic accidents, air pollution, fuel loss, and public dissatisfaction. Therefore, implementing traffic control systems that improve traffic flow and reduce travel times becomes essential. This work conducts a systematic literature review to identify the most efficient methods, algorithms, and models for developing traffic control systems. The review identifies three methods and three algorithms that are highly efficient for these systems, highlighting Bayesian filters and convolutional neural networks. It also shows that You Only Look Once (YOLO) is the most efficient image processing model for these implementations.