ExploDTwin

Simulation technology has been a cornerstone in the European Space Agency (ESA) and the broader European Space industry for several decades, playing a pivotal role across various phases of a space mission's lifecycle. The Digital Twin (DT) concept enhances simulators with dynamic and self-evolving models, with real-time bi-directional data assimilation, with the objective to support monitoring and control of space assets by creating a harmonious convergence between the physical reality and its virtual counterpart. The introduction of Artificial Intelligence (AI) in DTs enables data driven and symbolic reasoning for state synchronization, model adaption, planning, monitoring, diagnosis and prognosis. Intelligent DTs with AI are particularly relevant in scenarios where the operational environment undergoes dynamic changes, such as the exploration of new planetary areas. However, the integration of data analysis techniques and model-based symbolic reasoning to support the real-time space asset operations is very challenging due to the heterogeneity of the underlying models.

ExploDTwin ("Digital Twin for Space Exploration Assets") is a Technology Development activity carried out by TRASYS, FBK, and ALTEC with the support of the European Space Operations Centre (ESOC). The objective is to integrate the DT into the MOI to support online space assets operations with functionalities such as planning, what-if-analysis, fault detection, diagnosis and prognosis. To this end, the primary aim is to introduce a model-based design methodology that allows to seamlessly integrate engineering methods and Artificial Intelligence (AI) techniques into a cohesive Intelligent DT framework. Selected scenarios will be explored for the key use cases under investigation: state synchronisation, diagnosis of unexpected events and mission planning. A key aspect is the distinction between the DT infrastructure and the DT of the real asset for the specific use case. Infrastructure generic components such as the ‘DT Manager’, the ‘State Synchronisation Manager’, the ‘Simulation As a Service’, the ‘Model As a Service’ are tightly coupled with an MLOps environment. The design of the specific DT is approached with a formal model that contains the formal specification of the functionalities needed by the specific use case. Selected scenarios will be explored for the key use cases under investigation: state synchronisation, diagnosis of unexpected events and mission planning. The solutions will be demonstrated on both ESOC and ALTEC use cases.

General Info

Start Date: March 1, 2024
End Date: May 31, 2025
Duration: 16 months
Funding: ESA

Partners

• TRASY (prime)
• FBK participates with three units
  • FM
  • DSIP
  • PSO
• ALTEC

Contacts

Stefano Tonetta tonettas@fbk.eu