Integration of Cloud Computing with Internet of Things

7Traffic Prediction Using Machine Learning and IoT

Daksh Pratap Singh and Dolly Sharma*

Department of Computer Science and Engineering, Amity School of Engineering and Technology, Amity University, Noida, India

Abstract

Real-time traffic information uses advanced APIs and IOT-based mobile phone sensors to send information required like speed, average road density and differentiation using colors and graphs from low, moderate, and high traffic with a traffic simulator for creating customized routes to predict certain traffic scenarios manually. Google maps was very able to predict the traffic but was dependent only on GPS sensors to predict the traffic whereas this application stands out to be more precise by using a vehicle’s speed to predict the traffic. The objective of this work is to develop an android application by targeting user’s need to see real-time traffic using IOT, Machine Learning and GPS-based advance APIs. APIs are collecting data using crowdsourcing and real-time user’s location accumulation on server by microservices. This application consists of industry-ready MVVM architecture, JSON parsing, real-time data rendering, data manipulation, and IOT-based cloud computing. The application predicts traffic with utmost precision and is capable of taking a vehicle’s speed into consideration to predict vehicular traffic and is also capable of simulating the traffic on one’s own terms.

Keywords: APIs, GPS, machine learning, IOT, simulation

7.1 Introduction

The application ...