ETSI MEC standard compliant AI-driven cross-platform digital twin collaboration gateway for industrial metaverse

Abstract

This thesis presents the design and implementation of an ETSI-compliant, AI-driven digital twin gateway optimized for real-time 3D collaboration within distributed Multi-access Edge Computing (MEC) environments. We propose a modular, open-source design for low-latency and AI-facilitated service orchestration in response to the increasing demand for industrial metaverses and collaborative 3D applications that require low-latency, high-bandwidth processing. The architecture consists of three major components: MEC node, deployed in Germany, supporting service registration, liveness probes, and application context creation following ETSI MEC standards—the Main server deployed in Germany, routing users to specific nodes and storing essential data including 3D files and clients information, and two end users working in Finland and the USA. For the distributed processing of 3D content (e.g., FBX, USD, glTF formats), the platform provides services such as AI-based upscalers and format converters, all of which work under applications deployed on MEC nodes. This work presents a scalable framework for concurrent edge rendering and intelligent 3D asset management, which can be potentially applied to distributed metaverses, digital twins, and collaborative designs. The system's open nature also offers an opportunity for further exploration in context-aware service steering, heterogeneously interoperable 3D engines, and latency-constrained AI orchestration on subsequent MEC-supported metaverse platforms.