Platooning of Autonomous Vehicles with Artificial Intelligence V2I Communications and Navigation Algorithm

Abstract

A solution to provide greater road safety consists in making vehicles move much closer to each other, i.e. using platoons, and to avoid the constant stop-and-go in urban traffic. This paper discusses the implementation of platooning in Autonomous vehicles (AVs) and the different difficulties associated with it. It also proposes a physical platform able to perform a form of platooning using Artificial Intelligence (AI) to create platoons with miniature cars. The platform called Autonomous Learning Intelligent Vehicles Engineering-ALIVE is the association of multiple physical cars coupled with an infrastructure, which handles the data of every single car and makes decisions based on the augmented environmental perception. This augmented perception can be created thanks to the received data. Our platform is low-cost, energy efficient and easy to use for other researchers to test their platooning ideas or algorithms. Finally, we tested the platooning AVs under real conditions. Preliminary results demonstrate the effectiveness of ALIVE to create platoons between multiple cars, allowing a car to reach any destination without making any other decision than creating or entering a platoon with another car going to the same destination.