Autonomous Vehicle Control Using CARLA Simulator, ROS, and EPS HILS

Abstract

This paper proposes a method for integrating the electronic power steering (EPS) system into autonomous driving control by interfacing the CARLA simulator with the Robot Operating System (ROS). CARLA is a high-fidelity simulator designed for autonomous driving research, offering a realistic simulation environment for urban driving scenarios. In this study, ROS is utilized as middleware to facilitate seamless communication between the CARLA simulator and the EPS control system. The proposed system architecture enables realtime control of vehicle steering through EPS Hardware-In-the Loop-Simulation (HILS), supporting precise steering in autonomous driving tasks such as lane-keeping, obstacle avoidance, and path-following. The integration of these technologies demonstrates that CARLA and ROS can be effectively leveraged for the development and validation of autonomous driving algorithms. This study highlights the practical feasibility of EPS HILS-based steering control in a variety of driving conditions. Future work will focus on optimizing advanced control algorithms and expanding functionality by incorporating vehicle dynamics.