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

Abstract

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

Investigation of linear models for control of water flow and temperature in a water supply system

“…In this case, it is advisable to use PID controllers or their modification [19]. If successful, it will be possible to use water supply management using a machine-learning program [20] as well as adaptive control methods for a task with many objects [21]. It is also advisable to subsequently use Markov models of this system, as well as carry out mathematical modeling of the activity of object components in real time [22].…”

Distributed Multi-Intersection Traffic Flow Prediction using Deep Learning

“…Diverging from conventional shallow architectures, deep learning models employ multi-layer nonlinear structures to encapsulate distributed and hierarchical features within complex traffic flow datasets. Some researchers have introduced clustering methodologies, such as Deep Belief Networks (DBN), into the realm of traffic flow prediction [25]. Moreover, [26] have devised an intelligent swarm-based optimization framework for parameter tuning in DBN, augmenting its predictive prowess for multiple time steps ahead.…”

Elevating smart city mobility using RAE-LSTM fusion for next-gen traffic prediction

“…This work chronicles the methodical development of an intelligent short-and long-term urban traffic prediction system [14], [15], encompassing various mobility data types, traffic modeling techniques, and critical traffic indicators such as speed, flow, and accident risk [16], [17]. Special emphasis is placed on time series analysis and the pivotal role of data preprocessing, encompassing normalization, transformation, outlier handling, and feature engineering, crucial in enhancing the predictive accuracy of deep learning models [18]. Notably, LSTM demonstrates remarkable proficiency in handling prolonged time series data [19], [20].…”