Traffic Flow AI: Smart Traffic Prediction

Traffic Flow AI: Smart Traffic Prediction

Nirdesh Bhardwaj, Kritika, Vishav Pratap Singh, Rahul Kulvi

Department of Computer Science and Engineering Chandigarh University, India

Abstract

In the past few years, the rapid rates of autonomous vehicle consumption worldwide have been noticed. It`s because the growing use of AI technologies and AI systems in the market is often caused by rapid new technologies appearance, as well as brought about a great number of industries and organizations to take advantage of AI. Traffic flow predictability is a key point for autonomous vehicles they have a roadmap and make well-adaptive decisions (to move left or right, to drive straight, to change lanes, to stop or to accelerate) within the existing context. According to the available study, Adaptive machine learning technology is considered the main goal of autonomous car research, replacing statistical models with evolutionary algorithms. However since modern cars make their decisions based on their environment, the non-linear relationship between spatial and physical information could mean no. Yes, all modern learning systems are now suitable for this situation.

Therefore, we often cover these topics and examine the use of deep learning to predict the traffic of unmanned vehicles in the context of modern intelligent transportation (ITS). Many factors are particular to each model and are therefore selected to make problem-based comparisons among the deep learning models. Even more, we tried to show the difficulties and future research possibilities.