Scholarlink

The Industrial Metaverse extends Industry 4.0 by merging physical and virtual environments into an integrated platform for collaboration, simulation, and intelligent automation. Enabled by technologies such as Digital Twins, IIoT, AI, VR/AR, and photorealistic rendering, it offers new opportunities to accelerate innovation cycles, enhance sustainability and improve resilience in manufacturing. This chapter explores the technological foundations of the Industrial Metaverse and demonstrates its potential through the use of synthetic data for AI-based production systems. Two case studies on object recognition and automated quality inspection illustrate how simulation-based data generation and domain randomization address data scarcity and the Sim2Real gap, enabling more robust and cost-efficient AI applications. Although high implementation costs, integration challenges and user acceptance remain barriers, collaboration between research and industry shows promising pathways to overcome them. The Industrial Metaverse emerges as a disruptive enabler of future industrial production, driving digital transformation beyond current approaches.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Potentials and Applications of the Industrial Metaverse Using the Example of Synthetic Data Generation

  • 章节

  • 发表日期：

    2025年

查看全文
  • Oliver Petrovic,
  • Josefine Monnet,
  • Petar Tesic,
  • Yannick Dassen,
  • Werner Herfs

摘要

The Industrial Metaverse extends Industry 4.0 by merging physical and virtual environments into an integrated platform for collaboration, simulation, and intelligent automation. Enabled by technologies such as Digital Twins, IIoT, AI, VR/AR, and photorealistic rendering, it offers new opportunities to accelerate innovation cycles, enhance sustainability and improve resilience in manufacturing. This chapter explores the technological foundations of the Industrial Metaverse and demonstrates its potential through the use of synthetic data for AI-based production systems. Two case studies on object recognition and automated quality inspection illustrate how simulation-based data generation and domain randomization address data scarcity and the Sim2Real gap, enabling more robust and cost-efficient AI applications. Although high implementation costs, integration challenges and user acceptance remain barriers, collaboration between research and industry shows promising pathways to overcome them. The Industrial Metaverse emerges as a disruptive enabler of future industrial production, driving digital transformation beyond current approaches.