TEACar: An Open-Source Autonomous Driving Platform

2026-04-27

AI Summary

This work addresses the limitations of existing small-scale autonomous driving experimental platforms—such as insufficient modularity, complex integration, and restricted extensibility—which hinder hardware-in-the-loop validation of vision-based perception and learning control in intelligent transportation systems. To overcome these challenges, the authors propose an open-source 1/14–1/16 scale autonomous driving platform featuring a four-layer stacked mechanical architecture that physically decouples the perception, computation, actuation, and power subsystems. By integrating ROS 2 middleware with a hardware abstraction layer, the platform achieves a highly rigid, low-cost, and easily reconfigurable hardware-software system. Experimental results demonstrate that the platform, equipped with three CNN-based steering controllers, exhibits excellent mechanical stability, low inference latency, reasonable power consumption, and extended operational endurance, significantly enhancing its scalability and suitability for research and educational applications.

Abstract

Intelligent Transportation Systems (ITS) increasingly rely on vision-based perception and learning-based control, necessitating experimental platforms that support realistic hardware-in-the-loop validation. Small-scale platforms for autonomous racing offer a practical path to hardware validation, but often suffer from limited modularity, high integration complexity, or restricted extensibility. This paper presents TEACAR, a 1/14- to 1/16-scale autonomous driving platform designed with modular mechanical architecture, hardware abstraction, and ROS 2-based software. The system adopts a four-layer deck structure that physically decouples sensing, computation, actuation, and power subsystems, improving structural rigidity while simplifying reconfiguration. We constructed and comprehensively evaluated the prototype of TEACAR. Its mechanical stability, structural characteristics, and software performance were quantified based on three CNN-based steering controllers. Inference latency, power consumption, and system operating time were measured to evaluate computational capability and robustness. Our experiments demonstrated that TEACAR offers a scalable, modular, and cost-effective testbed for ITS research, education, and development. Our project repository is available on GitHub.

Problem

• autonomous driving platform
• modularity
• hardware-in-the-loop validation
• extensibility
• Intelligent Transportation Systems

Innovation

• modular architecture
• hardware-in-the-loop
• ROS 2
• autonomous racing platform
• hardware abstraction