Overview
By its very nature, a digital twin strongly relies on data visualization and user interactions. Indeed, a DT usually embeds interactive systems that enable users to visualize and interact with both the physical and digital twins and their large amount of associated data.
Those interactive systems can take multiple forms, from fully digital VR immersive systems to hybrid AR systems, but also 2D graphical user interfaces, and systems that imply physical input devices. They often propose collaborative asymmetrical (AV/VR/2D) interactive situations, which can also be a challenge.
Building such interactive systems from scratch or without automation is time-consuming and may prevent the generalization of DT as industrial tools. First, novel approaches are required to build the visual representation of the DT, which can be 3D models (to build, for example, by combining CAD models, 3D scans, pictures, and other media sources) or 2D various kinds of information. Users can then interactively finalize some parts of the DT. Additional visualization tools can also help reduce the bias of automatic reconstruction.
Then, DT come with usages that raise research questions regarding, from an HCI viewpoint, how to help users in visualizing data and interacting with it, especially when we use 3D. The visualization concerns not only the “raw” model of the physical twin but also all the data that can be used to enrich the model (additional information about how to use it). It raises new challenges about how to display large datasets of information and how and when to interact with them.
Finally, all these points require validations, at different scales. To reduce the cost of developing the HCI facets of DT, the project must propose novel techniques that focus on their engineering.
The project focuses on 3D reconstruction, data visualization, interaction engineering, spatio-temporal navigation, and validation with clear objectives: developing semi-automatic construction techniques, creating visualization metaphors, designing interaction approaches, implementing seamless transitions, and establishing usability assessment methods.
Associated Use Cases
Use cases in PC5 span multiple scales:
- Territory Scale: Reconstructing the DT of a city or area and populating it with virtual humans to better understand current limitations and how the DT can enhance it
- Building/Factory Scale: Reconstructing the DT of a reconfigurable assembly line, providing AR augmentation for workers and VR visualization for experts who can help remotely
- Technical Object Scale: Reconstructing the DT of a complex system (car, satellite) to train people for maintenance tasks using VR interactions, then using AR augmentations when physically in front of the physical twin
Investigator & Project Partners
Principal Investigators:
Thierry Duval
Full Professor and Head of Computer Science Department at IMT Atlantique
Thierry Duval is a full Professor and the head of the Computer Science Department at IMT Atlantique in Brest. He is member of the INUIT team at Lab-STICC. He received a PhD degree in Computer Science from Centrale Nantes and the University of Nantes in 1993, and an Habilitation degree from University of Rennes 1 in 2012. His research interests include Virtual, Augmented, eXtended and Mixed Reality, Immersive Analytics, Collaborative Virtual Environments, as well as Software Engineering for interactive systems. He is currently involved in several collaborative projects (5G Metaverse, Reclassif, eNSEMBLE) where he conducts AR/VR/XR/MR/CVE research mainly applied to Industry of the Future where Digital Twins of factories are central elements of the projects. He has published over 100 communications in international conferences and journals.
Arnaud Blouin
full Professor at the University of Rennes
Arnaud Blouin is a full Professor at the University of Rennes, France, where he conducts research within the KHORA research group. He conducts research in the field of software engineering, focusing on human-machine interaction engineering, software maintenance and validation, as well as domain-specific languages and their development environments. He currently co-leads the CNRS GDR-IHM working group on interactive systems engineering.
Participating Partners:
Project Implementation
The project is structured into three complementary work packages, addressing reconstruction and visualization of Digital Twins,their spatial and temporal exploration, and the modeling of their users interactions.
Workpackage1: DT Reconstruction and Visualization
Partners: Univ. Bourgogne, CEA LIST, Inria Aviz, Inria Seamless
Objectives:
- Develop automatic or semi-automatic techniques for 2D/3D construction of digital twin views and associated data
- Handle 3D reconstruction
- Manage 2D reconstruction
- Manage uncertainties and biases due to variability and liability of input data
Key Tasks:
- 3D reconstruction including object segmentation with kinematic and mechanical features (for rigid and articulated bodies)
- Bias and uncertainty estimation and visualization from Multi-source data integration and fusion
Workpackage2: DT Spatio-Temporal Exploration
Partners: Inria Seamless, IMT INUIT, Univ. Toulouse 3 ELIPSE, CEA-Tech Occitanie, CESI LINEACT Rouen, Université d’Évry IBISC
Objectives:
- Create exploration mechanisms to ease spatial and temporal exploration of the DT
- Ease AR/VR transitions from the physical world to its DT
Key Tasks:
- Capture and visualize temporal evolution od the DT and its interactions
- Allow parallel exploration of several DT clones resulting from different evolutions (resulting from different choices/decisions)
- Allow AR/VR transitions to switch easily from the physical world to its DT
Workpackage3: DT Operators Interactions
Partners: Univ. Toulouse Capitole ICS, CNAM CRTD, Inria DiverSE, UGA-INP G-SCOP, Bretagne INP Énib, CESI LINEACT Rouen, Inria Seamless
Objectives:
- Design approaches for describing and generating user interactions with DTs
- Implement interaction engineering including tangible user interfaces
- Enable co-engineering between multiple stakeholders
Key Tasks:
- Design interaction techniques and interaction devices
- Design tools for modelling interaction tasks
- Design tools to instantiate interaction scenarios
Related publications
Designing Usable User Interfaces of Digital Twins: Benefits and Limitations of Applying the User Centered Design Process
Pierre Treton, Hafiyyan Sayyid Fadhlillah, Camille Fayollas, Philippe Palanque, Arnaud Blouin, Benoit Combemale, Francois Fouquet • 2026
Conference Paper • HCSE 2026 - 11th International Working Conference on Human-Centered Software Engineering
Engineering Digital Twins: A Research Roadmap
Benoît Combemale, Pascale Vicat-Blanc, Arnaud Blouin, Hind Bril El Haouzi, Jean-Michel Bruel, Julien Deantoni, Thierry Duval, Sébastien Gérard, Jean-Marc Jézéquel • 2025
Conference Paper • EDTconf 2025 - 2nd International Conference on Engineering Digital Twins
Model‐Driven Engineering for Digital Twins: Opportunities and Challenges
Judith Michael, Loek Cleophas, Steffen Zschaler, Tony Clark, Benoît Combemale, Thomas Godfrey, Djamel Eddine Khelladi, Vinay Kulkarni, Daniel Lehner, Bernhard Rumpe, Manuel Wimmer, Andreas Wortmann, Shaukat Ali, Balbir Barn, Ion Barosan, Nelly Bencomo, Francis Bordeleau, Georg Grossmann, Gabor Karsai, Oliver Kopp, Bernhard Mitschang, Paula Muñoz Ariza, Alfonso Pierantonio, Fiona Polack, Matthias Riebisch, Holger Schlingloff, Markus Stumptner, Antonio Vallecillo, Mark van den Brand, Hans Vangheluwe • 2025
Journal Article • Systems Engineering
Related positions
Automatic construction of a digital model from real-world scenes
Extraction of the geometry, the kinematics, and the mechanical properties using machine learning of a single object from videos or photos.
- Master degree in relevant field
- Experience with digital twins
- +1 more requirements
Visualization of the Plausibility and Bias for Data Resources used in a Geographic Digital Twin
In this PhD research project we will investigate how to automatically classify the plausibility in datasets collected for a geographic or otherwise spatial data, and how to visualize it. We will rely on past results that detected data biases and data errors based on a manual analysis, and will investigate how to use machine learning models to automate this process. In addition, we will investigate how best to depict the different types and degrees of data bias, data imprecision, and data error.
- Master degree in computer science or related field
- Interest in interactive data visualization and analysis
- +4 more requirements
Supporting transitions for interacting in Digital Twins.
We are seeking a motivated PhD candidate to join the PC5 team working on Human-Digital Twin Interactions. This position focuses on developing innovative immersive solutions to support transitions between the physical referent (PR) and the digital replica (DR) of a Digital Twin (DT).
- Master degree in HCI, VR, Computer Science or other relevant fields
- Experience with AR / VR technologies (software, hardware)
- +2 more requirements
Ongoing Research Positions
Navigating through the temporal aspects of the digital twin
PhD position in Human-Computer Interaction and Virtual Reality focusing on immersive interfaces for digital twin systems
Exploring Alternative Evolutions of Digital Twins
PhD position in Human-Computer Interaction and Virtual Reality focusing on immersive interfaces for digital twin systems
A software engineering approach to ease the development and to ensure the usability of interactive digital twins
By its very nature, a digital twin strongly relies on data visualization and user interactions. Building such interactive digital twins from scratch and in a craft way is time-consuming and may prevent the generalization of DT as industrial tools limiting the benefits to be expected from their deployment. The goal of the PhD is to propose novel software engineering approaches to ease and support the development of reliable and usable interactive digital twins
Modeling Interaction Scenarios for Digital Twins in Mixed Reality
The effective use of digital twins requires interactive scenarios capable of organizing the system’s objectives, roles, activities, and behaviors within a real-time framework. This requirement is further heightened in mixed reality, where the user interacts with both digital representations and a physical environment, necessitating the joint design of the business model, interaction, and scenario execution.