Design and verification of an adaptive control system for small satellites involved in rendez-vous and docking missions.

As the title of this dissertation may suggests, the objective is to design a control system whom is able to perform a successful operation of autonomous rendez-vous and docking between two small satellites. This type of mission is becoming increasingly used to complete various tasks like removal of space debris, on-orbit servicing and spatial repairing. In order to control the Chaser satellite it has been developed an artificial neural networks (ANN), also known as “neural networks”, which are an useful mathematic/computational tool for the resolution of problems regarding control, data analysis and pattern recognition. In aerospace field, they are a strong candidate for attitude control thanks to its inherent nonlinear behaviour, which makes them a natural choice in order to control nonlinear system. The results are compared to the ones coming from a common-use controller, the Linear Quadratic Control (LQR).