Presentation Information
[GdsyDt6e-05]The Marine Autonomous Systems Digital Twin (MAS-DT) - Using AI to aid in the piloting of ocean gliders
*Justin Buck1 (1. National Oceanography Centre (UK))
The Marine Autonomous Systems Digital Twin (MAS-DT) project aimed to enhance ocean observation and forecasting by integrating digital twin technology with ocean modelling and marine autonomous systems. This approach demonstrated how a dynamic digital framework could optimise data collection and improve ocean model accuracy, delivering benefits for environmental science and operational forecasting.
At the core of the project was the development of a prototype digital twin to optimise ocean glider observations used by the UK Met Office. The system brought together multiple components: ocean glider operations from the National Oceanography Centre (NOC) under the Met Office Glider (MOGli) programme; earth observation data; and outputs from operational ocean models. These inputs were integrated within the NOC Command and Control (C2) infrastructure to create a four-dimensional representation of the ocean. A key feature was the Autonomy Engine developed by the Oxford Robotics Institute, which optimised glider missions based on defined objectives and constraints such as battery life and speed. Supporting this, an Objective Generation User Interface enabled scientists to identify high-impact observation areas and define mission goals.
Stakeholder engagement played a central role, including a joint workshop with the Turing Research Innovation Cluster for Digital Twins (TRIC-DT), which explored interoperability, user interaction, security, and system resilience. Enhancing interoperability across environmental digital twins was identified as an ongoing priority.
Overall, the project successfully demonstrated a functional digital twin integrating marine glider operations with the AMM15 ocean model. It delivered key innovations such as a new user interface and integrated mission planning tools. However, challenges included defining effective in-water experiments, accounting for human factors, and improving understanding of data assimilation processes. These findings highlight opportunities for future development, particularly in adaptive observation systems.
At the core of the project was the development of a prototype digital twin to optimise ocean glider observations used by the UK Met Office. The system brought together multiple components: ocean glider operations from the National Oceanography Centre (NOC) under the Met Office Glider (MOGli) programme; earth observation data; and outputs from operational ocean models. These inputs were integrated within the NOC Command and Control (C2) infrastructure to create a four-dimensional representation of the ocean. A key feature was the Autonomy Engine developed by the Oxford Robotics Institute, which optimised glider missions based on defined objectives and constraints such as battery life and speed. Supporting this, an Objective Generation User Interface enabled scientists to identify high-impact observation areas and define mission goals.
Stakeholder engagement played a central role, including a joint workshop with the Turing Research Innovation Cluster for Digital Twins (TRIC-DT), which explored interoperability, user interaction, security, and system resilience. Enhancing interoperability across environmental digital twins was identified as an ongoing priority.
Overall, the project successfully demonstrated a functional digital twin integrating marine glider operations with the AMM15 ocean model. It delivered key innovations such as a new user interface and integrated mission planning tools. However, challenges included defining effective in-water experiments, accounting for human factors, and improving understanding of data assimilation processes. These findings highlight opportunities for future development, particularly in adaptive observation systems.
Comment
To browse or post comments, you must log in.Log in