Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DMAVA: Distributed Multi-Autonomous Vehicle Architecture Using Autoware

Zubair Islam, Mohamed El-Darieby

TL;DR

DMAVA tackles the challenge of validating coordinated multi-AV behavior by introducing a distributed, real-time simulation framework that runs independent Autoware stacks across multiple physical hosts. It combines Autoware Universe, AWSIM Labs, and Zenoh to achieve synchronized inter-host communication with namespace isolation, enabling scalable multi-vehicle experiments without altering Autoware internals. Experimental validation in two- and three-host configurations demonstrates stable localization, reliable cross-host messaging, and synchronized closed-loop control, while identifying network timing as a key factor in scalability limits. The work provides a reusable infrastructure foundation for distributed multi-AV research and lays the groundwork for higher-level cooperative autonomy applications such as Multi-Vehicle Autonomous Valet Parking (DMV-AVP).

Abstract

Simulating and validating coordination among multiple autonomous vehicles (AVs) is a challenging task as most existing simulation architectures are limited to single-vehicle operation or rely on centralized control. This paper presents a Distributed Multi-AV Architecture (DMAVA) that enables synchronized, real-time autonomous driving simulation across multiple physical hosts. Each vehicle runs its own complete AV stack and operates independently from other AVs. The vehicles in the simulation maintain synchronized coordination through a low-latency data-centric communication layer. The proposed system integrates ROS 2 Humble, Autoware Universe, AWSIM Labs, and Zenoh to support concurrent execution of multiple Autoware stacks within a shared Unity-based environment. Experiments conducted on multiple-host configurations demonstrate stable localization, reliable inter-host communication, and fully synchronized closed-loop control. The DMAVA also serves as a foundation for Multi-Vehicle Autonomous Valet Parking, demonstrating its extensibility toward higher-level cooperative autonomy. Demo videos and source code are available at: https://github.com/zubxxr/distributed-multi-autonomous-vehicle-architecture.

DMAVA: Distributed Multi-Autonomous Vehicle Architecture Using Autoware

TL;DR

DMAVA tackles the challenge of validating coordinated multi-AV behavior by introducing a distributed, real-time simulation framework that runs independent Autoware stacks across multiple physical hosts. It combines Autoware Universe, AWSIM Labs, and Zenoh to achieve synchronized inter-host communication with namespace isolation, enabling scalable multi-vehicle experiments without altering Autoware internals. Experimental validation in two- and three-host configurations demonstrates stable localization, reliable cross-host messaging, and synchronized closed-loop control, while identifying network timing as a key factor in scalability limits. The work provides a reusable infrastructure foundation for distributed multi-AV research and lays the groundwork for higher-level cooperative autonomy applications such as Multi-Vehicle Autonomous Valet Parking (DMV-AVP).

Abstract

Simulating and validating coordination among multiple autonomous vehicles (AVs) is a challenging task as most existing simulation architectures are limited to single-vehicle operation or rely on centralized control. This paper presents a Distributed Multi-AV Architecture (DMAVA) that enables synchronized, real-time autonomous driving simulation across multiple physical hosts. Each vehicle runs its own complete AV stack and operates independently from other AVs. The vehicles in the simulation maintain synchronized coordination through a low-latency data-centric communication layer. The proposed system integrates ROS 2 Humble, Autoware Universe, AWSIM Labs, and Zenoh to support concurrent execution of multiple Autoware stacks within a shared Unity-based environment. Experiments conducted on multiple-host configurations demonstrate stable localization, reliable inter-host communication, and fully synchronized closed-loop control. The DMAVA also serves as a foundation for Multi-Vehicle Autonomous Valet Parking, demonstrating its extensibility toward higher-level cooperative autonomy. Demo videos and source code are available at: https://github.com/zubxxr/distributed-multi-autonomous-vehicle-architecture.
Paper Structure (14 sections, 4 figures, 5 tables)

This paper contains 14 sections, 4 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. System Design and Architecture
  4. Workflow-Based Architecture
  5. System Workflow
  6. Zenoh Communication Design
  7. Autoware and Localization Pipeline
  8. Extending AWSIM Labs for Multi-Vehicle Simulation
  9. Experiments and Results
  10. Experimental Setup
  11. Localization and Synchronization Performance
  12. Distributed Validation and Scalability Testing
  13. Communication Latency and Timing Behavior
  14. Discussion and Conclusions

Figures (4)

  • Figure 1: Proposed DMAVA showing the five-workflow design.
  • Figure 2: System-level workflow of the DMAVA across two physical hosts.
  • Figure 3: Zenoh-based communication flow across two hosts.
  • Figure 4: Consolidated visualization of two-host and three-host validation, demonstrating synchronized topic exchange and namespace isolation across distributed Autoware systems.