Space robotics require high-precision autonomous control technologies to meet the demands of On-Orbit Servicing (OOS) applications. In such scenarios, robotic manipulators must address the challenge of managing flexible vibrations in large and lightweight structures, which significantly impact control performance. Given the high level of difficulty associated with experimental testing and validating control approaches, this paper introduces recent advancements in a real-time simulation platform. This platform is designed to facilitate rapid conception, prototyping, and testing of model-based control solutions at a low time-scale. The software/hardware architecture of the platform ensures tight and high-fidelity space robot dynamics and flexible structures, incorporating visual environment models for state observation, computer vision processing, virtual sensor data fusion, and full robot control. To showcase the utility of the platform, a robust joint-space control framework is introduced, implemented, and contrasted with a separated base/manipulators control approach for an OOS scenario. The control structure consists of a Nonlinear Dynamic Inversion (NDI) for system linearization and decoupling, together with a structured H ∞ controller to provide robustness against flexible disturbances, model uncertainties, and sensor noise.