Internal Research Fellow (PostDoc) in Machine Learning for Advanced Space Control Systems

Our team and mission

Within the Systems Department, the GNC, AOCS and Pointing Division is responsible for the following activities:
• Engineering and R&D of Attitude and Orbit Control Systems (AOCS) for satellites including Failure Detection Isolation & Recovery (FDIR);
• Engineering and R&D of Guidance, Navigation and Control (GNC) systems for space vehicles including interplanetary cruise, aero assistance, precision landing, ascent, rendezvous and docking, re-entry, formation flying and drag-free systems;
• R&D of advanced control, estimation & optimisation techniques and tools;
• Technology development for AOCS and GNC sensors;
• End-to-end engineering and R&D of high pointing accuracy systems;
• Development of control engineering standards in the framework of ECSS and ESA standardisation boards.

Division objectives of particular relevance to the proposed Research Fellowship activities are indicated in the first three bullet points above.

Field(s) of activity/research for the traineeship

Our team has a strong background in advanced modelling of dynamical systems, system identification, simulation technologies, advanced robust and embedded control. AI technologies are deployed to vision-based navigation applications and worst-case analysis, including multi-parameter optimisation for control design.

As a subsequent step we are interested in extending our field to the development of certifiable physics-based machine learning AOCS/GNC and modelling concepts relying on a strong background in dynamical systems, controls and numerical optimisation technologies.

If selected for this fellowship, you will thus have the opportunity to work on a variety of challenging space projects with an emphasis on implementing the AI developed for navigation, estimation and control concepts on physical space systems.

We will also place a great deal of emphasis on the modelling opportunities provided by machine learning to allow efficient simulation of complex dynamics governed by partial differential equations.
Areas of research will be tailored with the team in line with the Agency’s strategic directions and activities.
  
Your research tasks will comprise:
• Developing a computational infrastructure for the design and validation on advanced ML systems by working together with team members to maximise synergies and to capture the problem areas and group needs.
• Collaborating on the Division’s space technology projects, for example in orbital robotics and satellites (active debris removal, Earth observation, telecom, exploration, space transportation systems) as well as re-entry systems towards the design and implementation of novel machine learning navigation, control and failure detection concepts, possibly in the Division laboratory facilities.
• Contributing to the development of advanced benchmark problems that will serve as baseline and reference cases for various R&D projects. This will be done in collaboration with industry and academia.
• Maintaining close contact with the world’s leading research communities involved in machine learning for dynamical systems, system identification and control systems.

       
As Research Fellow, you will be responsible for: 
• disseminating results obtained via technical reporting and peer-reviewed journals;
• communicating, disseminating and providing in-house expertise;
• in a broader context outside of the ESA environment, disseminating results via colloquiums and webinars;
• participating, with the team, in the assessment of proposed space system concepts providing support/advice in the technical areas related to machine learning concepts of strategic interest to ESA;
• supporting the elaboration of a machine learning AOCS/GNC strategy;
• making full use of the extensive engineering expertise available within the Group as well as within ESTEC, in order to foster collaboration and expand our knowledge domain.

Technical competencies

Behavioural competencies

• Result Orientation　
• Operational Efficiency　
• Fostering Cooperation　
• Relationship Management　
• Continuous Improvement　
• Forward Thinking

Education

You must have recently completed, or be close to completing, a PhD in a related technical or scientific discipline, preferably in dynamical systems, control systems and machine learning.

Additional requirements

A background in robust and embedded control, aerospace systems, physical modelling, simulation as well as AI-related concepts, electrical engineering, computer vision, and mathematics and numerical optimisation.
Proficiency in MATLAB, Julia, Python programming languages and associated packages needed for optimisation, modelling, requirements and machine learning are considered an asset.
Good analytical skills and the ability to work in a multicultural environment in an autonomous manner. You should have a methodical approach to your work and the ability to clearly document, disseminate and communicate your work and results.
The working languages of the Agency are English and French. A good knowledge of one of these is required. Knowledge of another Member State language would be an asset.

Specificities
The position of Research Fellow within the TEC-SA GNC, AOCS and Pointing Division is similar to a standard academic post-doc placement, but with a few key differences:

1. Although you are not required to be involved in teaching, it is presumed that hands-on knowledge dissemination within the Division will take place via working groups.
2. You will be involved in mentoring PhD students, Young Graduate Trainees and interns (stagiaires) within the team.
3. TEC-SA is a diverse, changing and interdisciplinary research team forming part of a large space agency, in contrast to a more specialised, focused research group with related or similar competences.
4.  TEC-SA aims to maximise contact with other disciplines, contributing to interdisciplinary research and collaboration.
5.  TEC-SA aims to communicate expertise and research results internally and externally, including the potential implications and importance for ESA’s long-term strategy.