Modelling and simulation of an ecological route in smart cities towards a green trajectory

The aim of this article is to develop a digital ecosystem of collective intelligence that will allow a resident to move around by taking a route based on several ecological metrics, such as traffic congestion, CO2 saturated areas and data retrieved in real time by city sensors, vehicles and connected object users [1]. We divide our research objective into three sub-problems: How to use the technological infrastructures of the Smart City to better store, transmit and manage CO2 emissions from road traffic? How to map the city in order to delimit the rate of CO2 emission and saturation of the different zones linked to road traffic? how to predict the CO2 level in a city? To this end, we propose a traffic flow prediction framework based on the hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model that takes into account the dynamic spatio-temporal characteristics of historical traffic conditions. Our experimental field is the city of Paris.