Preliminary design of a Floating Spacecraft Simulator for a Hardware-In-the-Loop testbed with a matching plane to simulate reduced gravity conditions

Simulating maneuvers in microgravity is challenging due to the unique conditions given by the space environment, which are difficult to replicate on Earth. These simulations are vital for developing operational protocols and proximity maneuver algorithms, helping to identify challenges, such as the precise control of the spacecraft during the maneuvers and to optimize algorithms, in terms of reducing time and power, to adjust the maneuver before the integration on the real spacecraft. As space missions become more complex, improving the fidelity and reliability of these simulations is crucial to ensuring mission success. The objective of this thesis is to delve into the most commonly used technologies that can replicate some conditions of a microgravity environment, that are the parabolic flights, the drop towers and the Floating Spacecraft Simulators(FSS) and then to analyze more deeply the FSS coupled with a granite matching plane. First, the thesis will focus on the design process of the FSS using Systems Engineering (SE) principles and elements identifying system requirements through SE tools such as the functional tree, the N2 matrix and the functions/products matrix. With this structured approach, it is possible, using the Analytic Hierarchy Process(APH) method, to select components that meet specific performance criteria, while also guaranteeing modularity and versatility in experimental configurations. The FSS incorporates AI-driven capabilities to enhance algorithm's training and test of autonomous systems in order to achieve real-time adjusted simulated maneuvers. AI integration is important because the implementation of AI-based techniques is increasingly recognized as a key enabler in space applications in the last years, with numerous studies exploring its potential to enhance spacecraft autonomy and control. Among these, Guidance and Control Neural Networks (G\&CNETs) have been proposed as a promising alternative to traditional onboard guidance and control systems, potentially replacing them with simplified neural models. Furthermore, reinforcement learning (RL) is being investigated as a method for empowering adaptive decision-making in unpredictable environments, such as autonomous navigation in space. One particularly relevant application involves the use of these techniques to maneuver spacecraft in non-uniform gravitational and rotational fields, where real-time adaptability is crucial for mission success. The implementation of the AI will be a crucial point for the next FSSs generation and in general for all the space applications involving the use of the GN\&C alghoritms. Then, the granite matching plane is taken into consideration mostly from a logistical perspective, with a preliminary plan of its installation inside a laboratory, due to the importance of a properly functioning testbed for the success of the project, ensuring the precision required for these applications. Trade-offs are assessed to determine the optimal installation position, considering the laboratory structural constraints and already present infrastructures layout. Subsequently, a preliminary installation plan is also presented from the unloading of the granite plane from its transportation vehicle to its desired positioning inside the laboratory by demolishing a drywall partition in order to facilitate the granite plane's entry and placement in the laboratory.