PhD studentship: AI based Multispectral Relative Navigation and Guidance and Control for Space Rendezvous and Docking Applications

This PhD studentship is offered in the Department of Engineering at City, University of London.

Project summary:

Active debris removal (ADR) and Autonomous On-Orbit Servicing (A-OOS) are of a proposed use of space rendezvous RV to mitigate space waste, a pressing issue jeopardizing current and future spacecraft launches. The proposed project aims to utilise the thermal imaging band, or long-wavelength infrared, as an alternative or complementary modality to the visible band to surmount the technical obstacles of vision-based spacecraft pose estimation with non-cooperative targets RVs and docking. Three-dimensional models of space objects to be used in artificial intelligence (AI) and machine learning (ML) are relatively simple to obtain in the visible; heat signatures on the other hand are hard to predict and can vary continuously throughout the orbit. Therefore, the large-scale learning capability of deep neural networks (DNNs) shall instead be leveraged to create models from restricted training sets capable of generalising to test sets containing unpredicted heating conditions of the target. The research shall include an investigation on how the developed model is taking its decisions in the context of eXplainable AI (XAI). This research project proposal will also aim at the creation of a Deep Reinforcement Learning (DRL) framework for the guidance and control task of the space vehicle in on-orbit docking applications considering the multispectral relative pose obtained. As such, this project proposes to combine classical control theory with DRL to reduce the learning cost and enable the transfer of taught behaviour from simulation to reality. This project offers the opportunity to collaborate with world-leading organisation like European Space Agency (ESA) through our contacts and research conducted with them on different programs of the team.

The candidate should have a minimum of an upper second-class BSc/BEng/MEng (or equivalent, or higher) degree in electrical engineering, mechanical engineering, aerospace/space engineering or computer science. Candidates should demonstrate expertise of conducting quality research related to the project aims and have a good background in Dynamics and Control (Space Dynamics is a plus), AI and Machine Learning, Image Processing, Programming (Python, C++, Matlab) and Hardware embedded implementation.

How to apply

Visit our group webpage www.ramigroup.co.uk for further information on the research we do. To make a formal PhD application please visit the link below: https://www.city.ac.uk/prospective-students/courses/research/electrical-and-electronic-engineering#accordion496142-header5 . You should enter the title of the research project as your proposal when applying.

The Scholarship is for 3 Years to start in February 2024.

This is a full-time position based in London at the main Campus of City, University of London.

Initial informal enquiries can be made to Professor Nabil Aouf at: [email protected] .

For further information:

https://www.city.ac.uk/prospective-students/finance/funding/smcse-doctoral-studentships

City, University of London is committed to promoting equality, diversity and inclusion in all its activities, processes, and culture for our whole community, including staff, students and visitors.

We welcome applications regardless of age, caring responsibilities, disability, gender identity, gender reassignment, marital status, nationality, pregnancy, race and ethnic origin, religion and belief, sex, sexual orientation and socio-economic background. City operates a guaranteed interview scheme for disabled applicants.