Artificial intelligence and machine learning (AI and ML) technologies have reached a state of maturity where they can help to substantially increase the level of automation in ground segment operations.
In this whitepaper, RHEA Group’s Pablo Beltrami and Nieves Salor Moral, with contributors from Airbus Defence and Space, Solenix and the European Space Agency’s (ESA’s) European Space Operations Centre, explain the outcomes of the AI4OPS project, which has been led by RHEA and designed to enable the adoption of AI and ML in mission operations.
Abstract introduction
Artificial intelligence/machine learning (AI/ML) technologies have reached a state of maturity where they can help to substantially increase the level of automation in ground segment operations. ESA/ESOC has established a multi-projects roadmap for the adoption of AI by mission operations teams, which includes establishing and making available the required infrastructure, including processes and reference information.
In this context, the AI4OPS project aims at the definition and implementation of a platform for developing and supporting AI-based applications targeting the automation of space operations and demonstrating it with a set of novel proof of Concept AI Applications.
The developed platform enables its users to build, train, and deploy ML models of AI applications, supporting mission operations. Additionally, the platform can also be used for general data engineering and software development purposes, allowing easy interface to data from ongoing or past space missions. The goal is the provision of a “platform as a service” (PaaS) available to internal ESOC projects and to data and software developments driven by ESA affiliates, contractors or other 3rd party collaborators.