Navigation and Guidance Algorithms for In-Orbit servicing Rendezvous Mission

The purpose of the thesis is to evaluate and validate Navigation and Guidance algorithms to perform autonomous In-Orbit Servicing (IOS) Space missions, a wide range of challenging new solutions for satellite operators. These missions address a large variety of useful applications based on Rendezvous and Proximity Operations (RPOs) technique, such as refuelling, life-extension, inspection, in-orbit assembly and active space debris removal. Particularly, the focus of this work is on position control and path planning during rendezvous and berthing operations, in which a Chaser satellite actively follows another satellite, the Target, and then joints with it using a robotic arm. In order to perform such manoeuvres, these algorithms must be robust and "flyable", which means that they should be able to effectively operate even with a system slightly different from the one expected and should have a low computational cost. The Navigation algorithms considered are based on the Sliding-Mode Observer (SMO) technique: the ability of observer-based algorithms to reconstruct the system dynamics despite receiving just the measurable states as input and the robustness of the Sliding-Mode technique have been the driving motors for this choice. The thesis also proposes an approach to perform data fusion using multiple position sensors, an optical camera and an accelerometer. Sensor simulation models have been developed in order to recreate realistic inputs for the Navigation algorithm, with different noise and sample rate. As for Guidance and Control, different approaches have been chosen depending on the distance of the Chaser from the Target. In the closing phase, two different strategies have been adopted and compared. In the first manoeuvre is designed a priori using Hill's equations and then executed in orbit exploiting a Sliding-Mode Controller (SMC). In the second strategy, the desired path is followed through the combined action of the Artificial Potential Fields (APF) method with moving goal and the Sliding-Mode Control (SMC). The latter strategy has been also adopted for the final approach, even though in this case the APF goal point is fixed and the path is entirely calculated online. The classic APF algorithm is able to attract the Chaser towards the Target performing obstacle avoidance considering both the Chaser satellite and the fixed obstacles as points. In this thesis the algorithm has been revised to contemplate the possibility of moving obstacles with their own geometrical shapes and take into account physical dimensions of the Chaser satellite. The major problem of APF, the presence of local minima, has been faced with the selection of harmonic 3D potential fields, characterised by the presence of global minima and maxima only, both located in the singular points at the end of these functions domain. Finally, the effectiveness of the selected Navigation and Guidance algorithms has been shown through several simulations conducted in MATLAB&Simulink environment, highlighting how the obstacle avoidance strategy impacts on the rendezvous trajectory and the propellant consumption and showing the performance of the combination of SMO and SMC in maintaining the desired final position during the motion of the robotic arm.