Adaptive traffic lights based on traffic flow prediction using machine learning models

<span lang="EN-US">Traffic congestion prediction is one of the essential components intelligent transport systems (ITS). This due to rapid growth population and, consequently, high number vehicles in cities. Nowadays, problem traffic attracts more and attention from researchers field ITS. Traffic can be predicted advance by analyzing flow data. In this article, we used machine learning algorithms such as linear regression, random forest regressor, decision tree gradient boosting K-neighbor regressor predict reduce at intersections. We public roads dataset UK national road test our models. All obtained good performance metrics, indicating that they are valid for implementation smart light systems. Next, implemented an adaptive system based on a model, which adjusts timing green red lights depending width, density, types vehicles, expected traffic. Simulations proposed show 30.8% reduction congestion, thus justifying its effectiveness interest deploying it regulate signaling intersections.</span>