Prevenção de acidentes em zonas rurais utilizando redes 5G e computação de borda

Abstract

Brazil's roads and highways are among the most dangerous in the world, mainly because most of them are not dual carriageways, and they are not well paved, and there is also a high incidence of potholes. Trucks are involved in almost half of the accidents, and this results in a large number of fatalities, mainly due to the structure and difference in mass between these vehicles and passenger vehicles. There are several risk factors for fatal accidents when there are a large number of trucks on the highways, and from this perspective, it is important to emphasize sugarcane trucks, as they have a reinforced structure in relation to other trucks, which ends up increasing the chances of fatalities in cases of accidents between them and smaller vehicles. Based on a system for sending alert messages, using the 5G network and edge computing, this work aimed to prevent accidents, mainly on roads and highways that are not dual carriageways, as they present a greater risk, but it was not limited to them. These two technologies were chosen as the basis for this work based on the observation of common problems that exist in rural and remote areas, such as highways, namely, high latency, low throughput, limited connectivity and even lack of internet connection. The use of the 5G network associated with edge computing may emerge as a potential solution to resolve or at least mitigate the problems mentioned above and this should enable the development of systems that can send and receive data in real time between vehicles, infrastructure and other devices on the network. The work indicated that it is extremely important to develop systems to prevent accidents on Brazilian highways, and based on a Systematic Review of the Literature, the main technologies capable of supporting the creation of such systems were observed. As a proof of concept, a simulator was used to demonstrate communication between vehicles in the most varied circumstances, given that it has an ecosystem with a preexisting network infrastructure, allowing interaction between its elements, in addition to a realistic virtual environment composed of different vehicles, cities and maps. The simulator also allows for the configuration of sensors, control and customization of the environment, simulation of different traffic situations, among others.