Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Miniature Testbed for Validating Multi-Agent Cooperative Autonomous Driving

Hyunchul Bae, Eunjae Lee, Jehyeop Han, Minhee Kang, Jaehyeon Kim, Junggeun Seo, Minkyun Noh, Heejin Ahn

TL;DR

This work tackles the validation gap in cooperative autonomous driving by introducing CIVAT, a 1:15-scale miniature testbed that couples miniature autonomous vehicles with smart infrastructure equipped with sensing and edge computing. The system enables real-time V2V and V2I communication via a Wi‑Fi/ROS2 framework, and supports both fully connected and mixed-traffic scenarios. It validates infrastructure-centric perception and intersection management through a multi-component software architecture and a case study, demonstrating real-time performance (average cycle ~31.8 ms) and high detection accuracy, while highlighting the complementary roles of infrastructure perception and V2I messaging. CIVAT offers a cost-effective, scalable platform for evaluating cooperative driving algorithms, paving the way for collaborative perception, infrastructure-assisted planning, and large-scale multi-agent coordination in practical settings.

Abstract

Cooperative autonomous driving, which extends vehicle autonomy by enabling real-time collaboration between vehicles and smart roadside infrastructure, remains a challenging yet essential problem. However, none of the existing testbeds employ smart infrastructure equipped with sensing, edge computing, and communication capabilities. To address this gap, we design and implement a 1:15-scale miniature testbed, CIVAT, for validating cooperative autonomous driving, consisting of a scaled urban map, autonomous vehicles with onboard sensors, and smart infrastructure. The proposed testbed integrates V2V and V2I communication with the publish-subscribe pattern through a shared Wi-Fi and ROS2 framework, enabling information exchange between vehicles and infrastructure to realize cooperative driving functionality. As a case study, we validate the system through infrastructure-based perception and intersection management experiments.

Miniature Testbed for Validating Multi-Agent Cooperative Autonomous Driving

TL;DR

This work tackles the validation gap in cooperative autonomous driving by introducing CIVAT, a 1:15-scale miniature testbed that couples miniature autonomous vehicles with smart infrastructure equipped with sensing and edge computing. The system enables real-time V2V and V2I communication via a Wi‑Fi/ROS2 framework, and supports both fully connected and mixed-traffic scenarios. It validates infrastructure-centric perception and intersection management through a multi-component software architecture and a case study, demonstrating real-time performance (average cycle ~31.8 ms) and high detection accuracy, while highlighting the complementary roles of infrastructure perception and V2I messaging. CIVAT offers a cost-effective, scalable platform for evaluating cooperative driving algorithms, paving the way for collaborative perception, infrastructure-assisted planning, and large-scale multi-agent coordination in practical settings.

Abstract

Cooperative autonomous driving, which extends vehicle autonomy by enabling real-time collaboration between vehicles and smart roadside infrastructure, remains a challenging yet essential problem. However, none of the existing testbeds employ smart infrastructure equipped with sensing, edge computing, and communication capabilities. To address this gap, we design and implement a 1:15-scale miniature testbed, CIVAT, for validating cooperative autonomous driving, consisting of a scaled urban map, autonomous vehicles with onboard sensors, and smart infrastructure. The proposed testbed integrates V2V and V2I communication with the publish-subscribe pattern through a shared Wi-Fi and ROS2 framework, enabling information exchange between vehicles and infrastructure to realize cooperative driving functionality. As a case study, we validate the system through infrastructure-based perception and intersection management experiments.

Paper Structure

This paper contains 30 sections, 1 equation, 10 figures, 5 tables, 2 algorithms.

Table of Contents

  1. INTRODUCTION
  2. TESTBED DESIGN
  3. Map Design
  4. Vehicle Design
  5. Hardware Components
  6. Kinematic Model
  7. Smart Infrastructure Design
  8. Communication
  9. Software Architecture
  10. CAV Architecture
  11. Planning
  12. Message Generator
  13. Infrastructure Architecture
  14. Perception
  15. Planning
  16. ...and 15 more sections

Figures (10)

  • Figure 1: Main Components of CIVAT
  • Figure 2: Road Types and Segments
  • Figure 3: Vehicle Components
  • Figure 4: Vehicle System Structure
  • Figure 5: Overall architecture of CIVAT
  • ...and 5 more figures