Digital twins for robot-assisted nuclear decommissioning: a review of perception, radiological mapping, and model interoperability, with the DORADO project as a case study

Nuclear decommissioning increasingly relies on digital technologies to support characterisation, planning, monitoring, and decision-making in environments where access is limited by radiation, contamination, or other operational constraints. Digital twins provide a framework for integrating geometric, radiological, and operational information, while robotic systems acquire data and perform tasks with reduced human exposure. Although often treated separately, the two are coupled through a continuous exchange of information: robotic observations update the digital representation of the facility, and the twin provides the context for mission planning and execution. This review examines that coupling from the perspective of information representation and information flow. Based on a structured search of the peer-reviewed literature, standards, and public technical reports, it covers geometric and radiological sensing, radiation-field estimation, geometric and semantic representations and their interoperability, geometry for radiation-transport calculation, localisation, change detection, mission planning, and data governance. Comparative tables summarise sensing technologies, radiological instruments, exchange formats, localisation approaches, international programmes, and the evidence obtained in real facilities to date. A dedicated section synthesises uncertainty propagation and verification and validation requirements for safety-relevant use. The EURATOM DORADO project is then presented separately as a case study, showing how the requirements identified in the review are addressed in an ongoing engineering programme and distinguishing published results from implementation and ongoing development. The review concludes that the main barriers are not the performance of individual sensors or algorithms but interoperability, end-to-end uncertainty management, validation under representative radiological conditions, and long-term data governance, and that most current systems operate at the level of digital shadows rather than fully bidirectional digital twins.