Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enabling Automated Integration Testing of Smart Farming Applications via Digital Twin Prototypes

Alexander Barbie, Wilhelm Hasselbring, Malte Hansen

TL;DR

The paper addresses the challenge of automated integration testing for smart farming software by introducing the Digital Twin Prototype (DTP), an emulation-based workflow that integrates sensor interfaces so emulated hardware matches real hardware inputs/outputs and supports continuous integration/delivery. Through a case study with SilageControl, the authors demonstrate how DTP, combined with ROS-based microservices and Gazebo simulations, enables development and testing without constant access to physical hardware, reducing costs and accelerating iteration, especially during off-season hardware shortages. Key contributions include a practical DTP implementation for agricultural robotics, an analysis of development and QA workflow challenges, and a roadmap for integrating simulations, synthetic data, and Docker-based CI/CD to improve reliability and speed in smart farming software. The work has practical impact by enabling hardware-agnostic testing, facilitating continuous deployment, and aligning with Industry 4.0 objectives in agriculture.

Abstract

Industry 4.0 represents a major technological shift that has the potential to transform the manufacturing industry, making it more efficient, productive, and sustainable. Smart farming is a concept that involves the use of advanced technologies to improve the efficiency and sustainability of agricultural practices. Industry 4.0 and smart farming are closely related, as many of the technologies used in smart farming are also used in Industry 4.0. Digital twins have the potential for cost-effective software development of such applications. With our Digital Twin Prototype approach, all sensor interfaces are integrated into the development process, and their inputs and outputs of the emulated hardware match those of the real hardware. The emulators respond to the same commands and return identically formatted data packages as their real counterparts, making the Digital Twin Prototype a valid source of a digital shadow, i.e. the Digital Twin Prototype is a prototype of the physical twin and can replace it for automated testing of the digital twin software. In this paper, we present a case study for employing our Digital Twin Prototype approach to automated testing of software for improving the making of silage with a smart farming application. Besides automated testing with continuous integration, we also discuss continuous deployment of modular Docker containers in this context.

Enabling Automated Integration Testing of Smart Farming Applications via Digital Twin Prototypes

TL;DR

The paper addresses the challenge of automated integration testing for smart farming software by introducing the Digital Twin Prototype (DTP), an emulation-based workflow that integrates sensor interfaces so emulated hardware matches real hardware inputs/outputs and supports continuous integration/delivery. Through a case study with SilageControl, the authors demonstrate how DTP, combined with ROS-based microservices and Gazebo simulations, enables development and testing without constant access to physical hardware, reducing costs and accelerating iteration, especially during off-season hardware shortages. Key contributions include a practical DTP implementation for agricultural robotics, an analysis of development and QA workflow challenges, and a roadmap for integrating simulations, synthetic data, and Docker-based CI/CD to improve reliability and speed in smart farming software. The work has practical impact by enabling hardware-agnostic testing, facilitating continuous deployment, and aligning with Industry 4.0 objectives in agriculture.

Abstract

Industry 4.0 represents a major technological shift that has the potential to transform the manufacturing industry, making it more efficient, productive, and sustainable. Smart farming is a concept that involves the use of advanced technologies to improve the efficiency and sustainability of agricultural practices. Industry 4.0 and smart farming are closely related, as many of the technologies used in smart farming are also used in Industry 4.0. Digital twins have the potential for cost-effective software development of such applications. With our Digital Twin Prototype approach, all sensor interfaces are integrated into the development process, and their inputs and outputs of the emulated hardware match those of the real hardware. The emulators respond to the same commands and return identically formatted data packages as their real counterparts, making the Digital Twin Prototype a valid source of a digital shadow, i.e. the Digital Twin Prototype is a prototype of the physical twin and can replace it for automated testing of the digital twin software. In this paper, we present a case study for employing our Digital Twin Prototype approach to automated testing of software for improving the making of silage with a smart farming application. Besides automated testing with continuous integration, we also discuss continuous deployment of modular Docker containers in this context.
Paper Structure (11 sections, 5 figures)

This paper contains 11 sections, 5 figures.

Table of Contents

  1. Introduction
  2. Digital Twin Prototypes
  3. Related and Previous Work
  4. SilageControl
  5. Research Design
  6. Results and Discussion
  7. The Challenges in the Development Workflow
  8. The Challenges for Product Customization and Quality Assurance
  9. A Digital Twin Prototype for SilageControl
  10. Threads to Validity
  11. Conclusion and Future Work

Figures (5)

  • Figure 1: Relationships of Digital Twin Prototypes with physical twins and digital twins
  • Figure 2: Realization and evaluation in ARCHES demomission
  • Figure 3: Sensor bar which monitors the process of silage making.
  • Figure 4: A simple digital model of a tractor mounted with the SilageControl sensor bar in a GAZEBO simulation.
  • Figure 5: Digital Twin Prototype of a tractor mounted with the SilageControl sensor bar in a GAZEBO simulation.

Theorems & Definitions (1)

  • Definition 1: Digital Twin Prototype