A HYBRID DEEP LEARNING MODEL FOR MANAGEMENT OF TRAFFIC

A HYBRID DEEP LEARNING MODEL FOR MANAGEMENT OF TRAFFIC

M.tech. Scholar Yash Sethi, Asst. Prof. Ajeet Singh Rajput
Department of Computer Science and Engineering

Rishiraj Institute of Technology (RIT),

Indore, India

Abstract: Some of the many uses of reliable traffic flow information include traffic forecasts, vehicle navigation devices, vehicle routing and congestion management. The management of traffic congestion is another example of an application. Unfortunately, because so few areas are equipped with sensors, getting real-time data on traffic flow is nearly impossible. Accidents and public events can have a significant impact on traffic flow and are difficult to foresee, making it difficult to predict traffic flow effectively. To make things more difficult, this is a factor. To begin, we'll utilise a dynamic traffic simulator to generate traffic on all of the links in the network using existing traffic data, demand predictions, and historical traffic data collected from sensors on the links themselves. To propose the use of hybrid deep learning to anticipate traffic flow on the road.
Keywords: Road Traffic, deep learning, Flow Prediction, Intelligent Transportation System