This thesis presents the design and experimental evaluation of an on-ground, low-cost docking simulation facility for testing Deep Reinforcement Learning (DRL) rendezvous and docking algorithms able to control small Cubesats. The facility is based on a repurposed 3D-printer, mechanically and electronically modified to execute pre-recorded trajectories generated by a numerical simulator in the loop with a DRL algorithm, scaled and fitted within the machine workspace. A dedicated embedded vision system acquires video of the entire execution for post-processing and comparison with the reference trajectories visualized in a SW Rendering engine. The motion system is driven by a Raspberry Pi 3B+ running Klipper firmware, while the vision system makes use of a Raspberry Pi 5 equipped with a Raspberry Pi Camera Module 3 Wide for visual data acquisition and processing.