A ROS-based Platform for Autonomous Vehicles: Foundations and Perception

Autonomous driving (AD) and Advanced Driver-Assistance Systems (ADAS) are at the forefront of today’s technical achievements in the field of automotive engineering, promising disruptive changes in the role of transportation in today’s society, and in the world’s economy in general. This endeavor is fueled by the production of new advances in research, and their swift and smooth transfer to the industry. The present work aims at helping with the latter: it’s a software and hardware platform based on ROS (Robot Operating System) that aims to allow researchers, students and companies in the field to share algorithms and data in the most painless way. It provides a functionality baseline that includes sensor support, localization, SLAM, object and lane detection, and basic navigation capabilities. Moreover, modern practices in distributed systems engineering are applied to allow users to easily monitor the system, diagnose problems, transfer and deploy the system’s components.