Program
The workshop will be a full day workshop on the 21st of March 2025
| Time | Details |
|---|---|
| 8:00-8:30 | Registration |
| Session 0 | Welcome Chair: Sara Montagna |
| 08:45-09:00 | Welcome Message from the Workshop Chairs: Sara Montagna, Samuele Burattini |
| Session 1 | Keynote Chair: Samuele Burattini |
| 9:00-10:00 | Digital Twins for Autonomous and Pervasive Computing Lukas Esterle |
| 10:00-10:30 | Coffee Break |
| Session 2 | Digital Twins Applications Chair: Samuele Burattini |
| 10:30-10:50 | Towards an Interoperable Digital Shadow Framework for Continuous Steel Casting using Explainable AI and Kubernetes Piotr Hajder, Mateusz Mojżeszko, Filip Hallo, et al. |
| 10:50-11:10 | Integrating Domain Adaptation and Causal Inference in Digital Twins for Plastic Injection Molding Gian Marco Paldino, Olivier Caelen, Marouene Oueslati, et al. |
| 11:10-11:30 | CONNECTED: A Knowledge Graph-Driven Platform for Clinical Data Harmonization and Personalized Digital Twin-Based Healthcare Alberto Marfoglia, Christian D’Errico, et al. |
| 11:30-11:50 | A Multi-Layer architecture for Human Digital Twin Francesco Franco, Lorenzo Lamazzi, Luca Bedogni |
| 12:00-13:30 | Lunch Break |
| Session 3 | Digital Twin Technologies Chair: Sara Montagna |
| 13:30-13:50 | Towards Multi-fidelity digital twins for dynamic adaption and efficient distributed simulation Mirgita Frasheri, Prasad Talasila, Peter Gorm Larsen and Lukas Esterle |
| 13:50-14:10 | AI-enabled Cybersecurity using Synthetic Data Fabrizio Baiardi, Salvatore Ruggieri, Vincenzo Sammartino |
| 14:10-14:30 | FIWARE as a Scalable Digital Twin Interface for DT Ecosystems: Benchmark and Limitations Kai Franke, Tobias Koch |
| 14:30-14:50 | XRTwin4Industry: A Comprehensive Framework for Standardization and Interoperability in XR-Enabled Industrial Digital Twins Timothé Lacheroy, Abdelhadi Belfadel, Jean-Yves Didier |
| 15:00-15:30 | Coffee Break |
| 15:00-15:20 | Event-Driven and Scalable Digital Twin System for Real-Time Non-Destructive Testing in Industrial Computational Systems Piotr Hajder, Mateusz Mojżeszko, Filip Hallo, et al. |
| 15:20-16:00 | Closing Session Chair: Sara Montagna, Samuele Burattini |
Keynote: Digital Twins for Autonomous and Pervasive Computing
Speaker: Lukas Esterle
Keynote Speaker: Lukas Esterle. Credits: Josefine Aaen Thygesen, AU Foto
Lukas Esterle is Associate Professor in Electrical and Computer Engineering at Aarhus University, Denmark, where he leads the research group on Autonomous Intelligent Systems. Lukas received his PhD with distinction (Dr. techn.) from Klagenfurt University, Austria, where he worked on selforganising networks and autonomous smart-camera systems. Afterwards he was a postdoctoral researcher at TU Wien, Austria, where he explored cyber-physical systems and worked on autonomous and selforganised planning and online modelling of multi-robot systems. During this time, he won a prestigious Marie-Skłodowska Curie fellowship funded by the European Commission and joined Aston University, UK, deepening his work on autonomously collaborating systems. In 2019, he joined Aarhus University, Denmark, and started work on digital twins and how to utilise them in autonomous systems, multi-robot applications, and pervasive computing. His research interests include distributed systems, multi-agent systems, and cyber-physical systems. He has published over 100 peer-reviewed papers in these areas and has been involved in several EU and national projects. He is also a member of the steering committee of the IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS) and an associate editor for the Journal of Ambient Intelligence and Smart Environments (JAISE) and ACM Transactions on Adaptive and Autonomous Systems (TAAS).
Abstract
Models are a fundamental part of operating autonomous, pervasive computing systems enabling such systems to have an understanding of their environment and themselves. Over the past years, Digital Twins (DTs), as an extension of simple models, have received significant attention from industry and academia alike. Over the past years, DTs are utilised in system design, development, and operation. Furthermore, due to their inherent ability to update and adapt to the real-world, they are excellent candidates to be used in autonomous, pervasive computing systems. In this keynote, I will outline challenges and initial solutions to utilise DTs in autonomous and pervasive computing systems. I will highlight challenges in the communication between DTs and their physical counterparts and the problems arising from asynchronicity and how to deal with these challenges. Due to the dynamics of the real-world, in which modern computing systems are deployed, not all aspects might be known at design-time and require adaptations of the DTs during runtime in order for the system to operate as expected. This brings challenges not only due to how to update them safely but also when to perform such updates without the intervention of an operator. However, before updating models and bringing them in operation, these models need to be tested and evaluated thoroughly in order to avoid failures during operation. I will discuss how to utilise DTs to explore problems by performing fault injection on DT models. Finally, as modern computing systems are not operating in isolation, I will also present the potential of DTs in collaborative systems and utilising them for collectives. I will conclude this talk with in an outlook into future research and specifically the planned work in the DT-CORE project.