Enhancing Autonomous Vehicles Control: Distributed Microservices with V2X Integration and Perception Modules

Abstract

This paper proposes an architecture and a real deployment for the shared perception between a vehicle and the infrastructure in autonomous mobility, employing microservices on the infrastructure and vehicle platforms to integrate data from multiple sensors and distributed computing units. Using V2X communication, the system autonomously controls vehicles in critical emergency scenarios, such as a vehicle over the presence of Vulnerable Road Users (VRUs), aided by a road side node. The integration with the Autoware platform extends control to the vehicle’s units and powertrain. With edge computing, rapid object detection via computer vision, and V2X message exchange, the proposed system ensures real-time integration with the vehicle’s autonomous control. The results of the autonomous vehicle in the presence of a VRU demonstrate the capability to deliver the alert to the vehicle within a valid time and the successful reaction by the vehicle. To the best of the authors’ knowledge, this marks the first integration of the control of an autonomous vehicle platform over alerts produced by internal/external sensors in a real scenario using real-time data, and through V2X communications.