A short-term traffic flow prediction based on recurrent neural networks for road transportation control in ITS

Abstract

Background/Objectives: In order to overcome the rising issue of traffic congestion, effective and exact traffic flow information is needed. Though, there have been lots of research and work being done on traffic predictions; still this field of research needs more attention. Methods/Statistical analysis: In this work, we have collected some real time traffic data for analyzing different flow patterns under different environmental conditions. In this paper, we present a short-term traffic flow prediction using RNN (Recurrent Neural Networks) for Road Transportation Control in ITS. Findings: Prediction of accurate traffic rate flow at any given time interval which is of vital importance in assisting and managing the road traffic conditions in smart cities. Improvements/Applications: Applied system uses deep learning technique for accurate predictions of traffic flow rate at any specific time.