³ÉÈË´óƬ

During his time as internal research fellow at the European Space Agency, Marcus Märtens organised several open competitions that paved the way for applying modern machine learning (ML) techniques to space-related challenges. This talk provided an overview of this process from the original idea to the preparation of the datasets, the management of the actual competitions, and an analysis of the resulting aftermath.

A particular focus is on the generation of a ML ready dataset that minimises the need for specific domain knowledge in order to attract a wide range of competencies and expertise. By having scientific groups from all over the world compete with each other, effective benchmarks are created and maintained even long after the competitions have ended. On the technical side, this talk focused on the super-resolution of Earth images, pose estimation of satellites, and the optimisation of orbital megastructures like Dyson rings.