Traffic Prediction for Intelligent Transportation Systems Using Machine Learning

Abstract

The goal of this project is to provide a platform for forecasting accurate and timely traffic data. Traffic conditions include things that can affect road traffic speeds, such as: B. Traffic lights, accidents, protests and even road repairs that can cause traffic jams. Motorists or drivers should make informed decisions when they have very accurate prior knowledge of all of the above approximations and more real-world conditions that may affect traffic. I can. can be lowered. It is also useful for the development of self-driving cars. Transportation data has increased dramatically over the past decades and is evolving towards the concept of transportation big data. Current traffic prediction approaches use specific traffic prediction models that are still inadequate to handle real-world situations. Therefore, we tackled the problem of traffic prediction using traffic data and models. Due to the vast amount of data available in transportation systems, it is difficult to accurately predict traffic flows. In this study, we wanted to use machine learning, genetics, soft computing, and deep learning techniques to evaluate vast amounts of data in transportation systems while greatly reducing complexity. In addition, it uses image processing algorithms to recognize traffic signs and ultimately help train self-driving cars properly.