Improved Arithmetic Optimization with Deep Learning Driven Traffic Congestion Control for Intelligent Transportation Systems in Smart Cities

Abstract

Abstract In the last few years, some progress had been made in smart cities, and reduction in traffic congestion was the topmost concern in the development of smart cities. Shorter delays in transmission between Roadside Units (RSUs) and vehicles, road safety, and smooth traffic flow are the major difficulties of Intelligent Transportation Systems (ITS). The rapid improvement in automobiles occurs which increased the number of road accidents and traffic congestion. Machine Learning (ML) was an advanced technique to find hidden insights into ITSs without being explicitly programmed by learning from datasets. This article introduces an Improved Arithmetic Optimization with Deep Learning Driven Traffic Congestion Control (IAOADL-TCC) for ITS in Smart Cities. The presented IAOADL-TCC model enables traffic data collection and route traffic on existing routes for avoiding traffic congestion in smart cities. To accomplish this, the IAOADL-TCC model employs hybrid convolution neural network attention long short-term memory (HCNN-ALSTM) method for traffic congestion control. In addition, IAOA based hyperparameter tuning strategy is derived to optimally modify the hyperparameter values of the HCNN-ALSTM model. The presented IAOADL-TCC model effectively enhances the flow of traffic and reduces congestion. The experimental result study of the IAOADL-TCC method can be tested by making use of road traffic dataset from Kaggle repository. The experimental outcome stated the better performance of the IAOADL-TCC model over other DL methods.