The advent of Industry 4.0 has boosted the usage of innovative technologies to promote the digital transformation of manufacturing realities, especially exploiting the possibilities offered by cyber physical systems and virtual environments (VEs). Digital Twins (DTs) have been widely adopted to virtually reproduce the physical world for training activities and simulations, and today they can also leverage on the integration of Machine Learning (ML), which is considered a relevant technology for industry 4.0. This paper investigates the usage of a combination of DT and ML technologies in the context of a real production environment, specifically on the creation of a DT enhanced with YOLO (You only look once), a state-of-the-art, real-time object detection algorithm. The ML system has been trained with synthetic data automatically generated and labelled and its performance enables its usage in the VE for real-time users training.
Machine Learning and Digital Twin for Production Line Simulation: A Real Use Case
01.01.2022
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
DDC: | 629 |
MACHINE LEARNING BASED DIGITAL TWIN SIMULATION AND PREDICTION OF STRUCTURAL COMPONENT'S FATIGUE LIFE
British Library Conference Proceedings | 2019
|A Digital Twin for Remote Learning: A Case Study
Springer Verlag | 2023
|Dynamic monitoring system of prefabricated component production line based on digital twin
British Library Conference Proceedings | 2022
|Digital Twin-based Flight Dynamics Simulation
IEEE | 2024
|