Placement project spotlight: Natalia lays the groundwork for space-based visual navigation at Craft Prospect

Earlier this year, Craft Prospect was delighted to welcome Natalia Bajnokova to the team through the University of Glasgow College of Science and Engineering Graduate School PhD Student Placement Programme. 

Working within our Responsive Operations team under the supervision of Luis Sánchez Fernández-Mellado, Natalia’s four-month placement, which wrapped up earlier this month, focused on a challenging and highly relevant area of autonomous space operations: relative pose estimation. The goal was to develop a machine-learning-based method that enables a spacecraft to estimate the relative position, orientation, and distance of another spacecraft using onboard imagery. 

This kind of capability is important for future missions involving rendezvous and close proximity operations, such as debris collision avoidance, inspection, and in-orbit servicing, where fast and reliable visual navigation can support safer and more autonomous decision-making. 

Building a visual navigation pipeline  

As part of the project, Natalia developed a complete relative pose estimation pipeline. The approach combined several tasks including the detection of an approaching spacecraft in an image, generation of a 3D model for this spacecraft, identification of key landmarks in an image, and the estimation of the final relative pose of the spacecraft by fitting the 3D model to these landmarks. 

Together, these steps formed a full end-to-end method for visual pose estimation from a single image, providing a strong foundation for future follow-on work in vision-based navigation and autonomous spacecraft operations. 

A successful placement project 

Natalia’s work delivered very promising results within the short timescale of the placement. In her own reflection on the project, she described it as a busy and rewarding four months, highlighting the pose estimation work as a particular success: 

‘We developed a complete pipeline capable of estimating the relative distance and orientation of an approaching spacecraft using only a single image. The results were very promising, and this work provides a strong foundation for future follow-on vision-based navigation projects.’ 

Projects like this are exactly why Craft Prospect values the University of Glasgow placement programme so highly. Each year, it gives us the opportunity to work closely with talented researchers on cutting-edge technologies and explore how advanced academic ideas can be translated into practical space capability. 

Natalia made a fantastic contribution during her time with us. Her placement has been a great example of what this programme can make possible: meaningful technical work, strong collaboration, and results that create a platform for future development.  

We’re excited to see where her next steps lead!