Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Toward Reproducibility of Digital Twin Research: Exemplified with the PiCar-X

Alexander Barbie, Wilhelm Hasselbring

TL;DR

The paper addresses the reproducibility gap in digital twin research by formalizing a comprehensive digital twin concept and demonstrating it end-to-end on a PiCar-X. It introduces a reproducible laboratory setup comprising the physical twin, a Gazebo-based digital model, a digital template, a digital thread, a digital shadow, a digital twin, and a digital twin prototype, all supported by open-source code and Docker-compose configurations. The work integrates MQTT and Apache Avro for efficient bidirectional data flow, and employs a multi-master ROS/GAZEBO architecture with emulators to enable Software-in-the-Loop testing within CI/CD pipelines. This approach lowers barriers to replication, enables automated testing across architectures, and supports rapid, agile development in embedded systems within Industry 4.0 contexts.

Abstract

Digital twins are becoming increasingly relevant in the Industrial Internet of Things and Industry 4.0, enhancing the capabilities and quality of various applications. However, the concept of \dts lacks a unified definition and faces validation challenges, partly due to the scarcity of reproducible modules or source codes in existing studies. While many applications are described in case studies, they often lack detailed, re-usable specifications for researchers and engineers. In previous research, we defined and formalized the \dt concept. This paper presents a reproducible laboratory experiment that demonstrates various \dt concepts. Our formalized concept encompasses the \pt, the digital model, the digital template, the digital thread, the digital shadow, the \dt, and the \dtp. We illustrate this series of concepts by using a PiCar-X, showcasing the progression from a \pt to its \dtp. The entire code base is published as open source, and for each concept, Docker-compose files are provided to facilitate independent exploration, understanding, and extension.

Toward Reproducibility of Digital Twin Research: Exemplified with the PiCar-X

TL;DR

The paper addresses the reproducibility gap in digital twin research by formalizing a comprehensive digital twin concept and demonstrating it end-to-end on a PiCar-X. It introduces a reproducible laboratory setup comprising the physical twin, a Gazebo-based digital model, a digital template, a digital thread, a digital shadow, a digital twin, and a digital twin prototype, all supported by open-source code and Docker-compose configurations. The work integrates MQTT and Apache Avro for efficient bidirectional data flow, and employs a multi-master ROS/GAZEBO architecture with emulators to enable Software-in-the-Loop testing within CI/CD pipelines. This approach lowers barriers to replication, enables automated testing across architectures, and supports rapid, agile development in embedded systems within Industry 4.0 contexts.

Abstract

Digital twins are becoming increasingly relevant in the Industrial Internet of Things and Industry 4.0, enhancing the capabilities and quality of various applications. However, the concept of \dts lacks a unified definition and faces validation challenges, partly due to the scarcity of reproducible modules or source codes in existing studies. While many applications are described in case studies, they often lack detailed, re-usable specifications for researchers and engineers. In previous research, we defined and formalized the \dt concept. This paper presents a reproducible laboratory experiment that demonstrates various \dt concepts. Our formalized concept encompasses the \pt, the digital model, the digital template, the digital thread, the digital shadow, the \dt, and the \dtp. We illustrate this series of concepts by using a PiCar-X, showcasing the progression from a \pt to its \dtp. The entire code base is published as open source, and for each concept, Docker-compose files are provided to facilitate independent exploration, understanding, and extension.
Paper Structure (11 sections, 10 figures)

This paper contains 11 sections, 10 figures.

Table of Contents

  1. Introduction
  2. Exemplifying of the Digital Twin Concept
  3. The Physical Twin
  4. The Digital Model
  5. The Digital Template
  6. The Digital Thread
  7. The Digital Shadow
  8. The Digital Twin
  9. The Digital Twin Prototype
  10. Digital Twin Prototypes in CI/CD-Pipelines
  11. Conclusion

Figures (10)

  • Figure 1: The physical twin of the PiCar-X.
  • Figure 2: Software running the PiCar-X physical twin.
  • Figure 3: The digital model of the PiCar-X used in the GAZEBO simulation.
  • Figure 4: The digital model of the PiCar-X, including the software component and the CAD model in a GAZEBO simulation.
  • Figure 5: The digital shadow of the PiCar-X. The communication from physical twin to the digital shadow is fully automated.
  • ...and 5 more figures

Theorems & Definitions (7)

  • Definition 1: Physical Twin
  • Definition 2: Digital Model
  • Definition 3: Digital Template
  • Definition 4: Digital Thread
  • Definition 5: Digital Shadow
  • Definition 6: Digital Twin
  • Definition 7: Digital Twin Prototype