Abstract: A large fraction of the information about galaxy formation is encoded in very faint stellar structures, such as diffuse galaxies, extended stellar halos, and low-mass satellites. These components are extremely difficult to observe, as they lie at surface brightness is well below the limits of most current surveys. The Arrakihs-F2 mission is being designed to specifically target this low surface brightness regime, providing deep optical and near-infrared observations of the nearby Universe. In this talk, I will present the most recent advances in the Arrakihs science case and on the design of the satellite and will also show how the high-resolution cosmological hydrodynamical simulations we are developing in MareNostrum5 under an EuropHPC Extreme Scale project will be used to interpret and support the science goals of Arrakihs. These simulations will allow us for the first time to connect the observed faint stellar light with the underlying distributions of stars, gas, and dark matter, and to quantify which types of objects and structures can realistically be detected. By combining Arrakihs observations with HARKONENS, we aim to place the low surface brightness Universe in a solid physical context and better constrain the processes that shape galaxies at the faint end.