CE10 - Industrie et usine du futur : Homme, organisation, technologies

Coherence of the digital twins for the industry of the future: Modeling, visualization and interaction with 4D digital mockups coupled with physical systems – COHERENCE4D


Coherence of the digital twins for the industry of the future: Modeling, visualization and interaction with 4D digital mockups coupled with physical systems

Context, motivation and objectives

Introduced at the beginning of the century by NASA, the concept of the digital twin is now at the heart of digitalization issues and appears to be an essential element for meeting the challenges of the fourth industrial revolution. The digital twin is a dynamic virtual representation that can follow the evolution of a physical product or system throughout its life cycle. It uses real-time digital data to enable understanding, learning and reasoning. Coupled with virtual reality (VR) and augmented reality (AR) systems as well as artificial intelligence (AI) algorithms, the digital twin allows to support decision-making processes more efficiently. <br /><br />To take full advantage of the possibilities of digital technology, the digital twin must be kept up to date over time and must reflect as closely as possible the state of its physical twin. Depending on the operating scenario, digital coherence maintenance must be performed synchronously and in real time, or asynchronously and staggered in time. In some cases, it can also be interesting to keep the history of the evolutions. In addition to the temporal dimension, maintaining coherence requires modifications at several levels: semantics, structure and geometry.<br /><br />The COHERENCE4D project aims to develop a new paradigm of modeling, visualization, interaction, and coherence maintenance of the digital twin interfaced to the physical system. Thus, the maintenance of coherence will be done in four dimensions, and thus in 4D, to take into account the spatio-temporal character of the evolutions: the digital twin modeled in three dimensions (3D) will adapt to follow the temporal evolutions (fourth dimension) of the physical twin. Depending on the scenario, the maintenance of coherence can be automated or require manual interventions through AR/VR interfaces. Between the real world and the virtual world, ad-hoc interfaces will allow the acquisition of information in order to restore it within the framework of operating scenarios, to allow decision making and to facilitate the implementation of actions, in particular via adapted visualization and interaction systems. Three main scenarios have been identified and will be used to validate the COHERENCE4D approach: «as-built/maintained«, «as-is/positioned« and «as-manufactured/aged«.<br /><br />The proposals made within the framework of the COHERENCE4D project will be validated on test cases and according to scenarios that will involve digital mock-ups of manufactured products and production systems. The requirements cover the entire life cycle with a particular focus on the operational phases.

The COHERENCE4D project is broken down into 5 main Work Packages (WP) which are articulated around the main objective of maintaining 4D coherence. The project will result in functional bricks and a demonstrator to validate, via proofs of concept (POCs), the models, methods and tools developed according to the following approach:
- WP1 - Digital twin and 4D coherence: Definition of the digital twin with time-varying geometry and topology and associated manipulation operators. The information related to the physical system will come from WP3 which will have preprocessed the output data of WP2. The digital models thus produced will have to be able to interface with the AR/VR tools as tested in WP4;
- WP2 - Multimodal data acquisition: Methodology for specification and integration of acquisition interfaces (e.g. scanners, cameras, trackers, sensors) and interfaces with APIs of physical systems and IIoT platforms. The acquired data (e.g. point clouds, images) will be exploited by the decision support tool of WP3 as well as by the coherence maintenance module of WP1;
- WP3 - Steering and decision support: Development of real/virtual comparison algorithms on multimodal data from WP1 and WP2, specification and experimentation of machine learning tools for the identification of update rules for the digital twin according to the evolutions of the physical twin, definition of estimators (e.g. estimation of the time to coherence). These algorithms will be integrated into the decision support module to allow automatic updating or via the interfaces of WP4;
- WP4 - Visualization and Interaction Metaphors: Methodology for the specification and integration of visualization and interaction interfaces for enhanced user experience and improved decision making. The interactions will have to be natural, whether in the field during the adaptations of the physical twin, or remotely during the exploitation of the digital twin. It will also be necessary to define new metaphors for the monitoring of temporal evolutions associated with the 4D digital models coming from WP1, and for the decision making in the framework of WP3;
- WP5 - Integration and validation: The aim here is to integrate in a demonstrator the functional bricks developed in WP1 to WP4. This demonstrator will then be tested via proofs of concept (POCs) which will allow the validation of the COHERENCE4D approach on each of the three identified scenarios. The scenarios will be based on case studies and data from previous work of the consortium members.

The results can be broken down into 4 levels that follow the breakdown of the WPs:
- Digital twin and consistency maintenance mechanism: the possibility of having a digital twin that is kept up to date with changes in its physical twin will have a major impact on many industrial processes. These advances will allow to use all the power of numerical simulation, optimization, and to take decisions quickly to be more agile, more precise and more autonomous. The results obtained will clearly allow to gain in competitiveness to meet the challenges of the industry of the future;
- Data acquisition and instrumentation of the physical twin: the methodology developed for the instrumentation of the physical twin will be a real breakthrough to support companies in the practical implementation of the digital twin for a greater digitalization of organizations. These results will make it possible to define packages that meet certain types of digital twins and allow for faster digitization of industrial processes;
- Decision support and knowledge capitalization: the mechanisms and algorithms that will be developed will have an impact on the automation of the decision-making chain. These functional blocks (e.g. AI-based consistency time estimators) can then be used in many other decision-making processes;
- Advanced visualization and interaction: the visualization and interaction metaphors, as well as the real/virtual synchronization tools, implemented in the framework of this project can be extended and/or adapted to other domains such as medical, industrial facilities monitoring, finance and more generally to all domains where this type of improvement can change the way of approaching a problem, a situation, a decision making to ultimately improve the realization of the associated tasks

The benefits will be on several levels. First of all, the results of the COHERENCE4D project will be of great interest to the manufacturing industry and to companies motivated by the deployment of the Industry and Factory of the Future via the digitalization of their systems and processes. The scenarios specified and validated in WP5 will serve as a springboard for direct and rapid implementation in companies with which the consortium members are in close contact.

All these perspectives demonstrate the impact that the results of the COHERENCE4D project can have at different levels of the industry, from large groups to SMEs. The consortium thus mobilized will be at the forefront of the development and concrete implementation of the Industry and Factory of the Future.

The results obtained will be disseminated through different channels (scientific publications, specialized journals, industrial expositions, website and social networks, etc.). The dissemination of scientific results will be done in international journals (Computer-Aided Design, Computers in Industry, Engineering with Computers, Virtual Reality, Artificial Intelligence for Engineering Design Analysis and Manufacturing, etc.), specialized journals (Arts et Métiers Magazine, Usine Nouvelle, etc.), as well as through communications in conferences and workshops with a national dimension (Colloque National S. mart, Journées du Groupe de Travail en Modélisation Géométrique des GDR IM et IG-RV, Journées thématiques du GdR IG-RV, etc.) and internationally (Solid and Physical Modelling - SPM, International Joint Conference on Mechanics, Design Engineering and Advanced Manufacturing - JCM, Computer-Aided Design and Applications - CAD, etc.). The publications will also be put on the multidisciplinary open archive HAL.

The COHERENCE4D project aims at developing a new paradigm for modeling, visualizing, interacting, and maintaining the coherence of 4D digital twins interfaced with physical systems that evolve over time. The work will focus on the definition of a model with variable geometry and topology over time, on the development of update mechanisms, on the definition of a new methodology for the specification and integration of acquisition interfaces, on the development of a decision support system exploiting an Artificial Intelligence capable to analyze the real/virtual gaps in order to decide on the updates to be made, on the specification and the experimentation of new metaphors for the visualization and interaction with the 4D digital twins through AR/VR devices. The building blocks will be integrated within a demonstrator and the overall approach will be validated through several proofs of concept (POCs) on scenarios related to the maintenance of manufactured products, the reconfiguration of production systems, and the quality control on production lines.

Project coordination

Jean-Philippe PERNOT (Ecole Nationale Supérieure d'Arts et Métiers - Laboratoire d'Ingénierie des Systèmes Physiques Et Numériques)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


Grenoble INP - G-SCOP Grenoble INP - Laboratoire des Sciences pour la Conception, l'Optimisation et la Production de Grenoble
ENSAM - LISPEN Ecole Nationale Supérieure d'Arts et Métiers - Laboratoire d'Ingénierie des Systèmes Physiques Et Numériques
UTC - Roberval Université de Technologie de Compiègne - Laboratoire Roberval. Unité de recherche en mécanique acoustique et matériaux.
AMU - LIS Aix-Marseille Université - Laboratoire d'Informatique et Systèmes

Help of the ANR 623,188 euros
Beginning and duration of the scientific project: - 42 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter