Development of an innovative algorithm (Huber-based Unscented Kalman Smoother) for autonomous satellite pose estimation in orbit for rendezvous and docking missions (tools used: MATLAB and Simulink). The thesis objective was to tackle space mission related issues in pose estimation. In particular, I faced the challenges to model data outliers, to handle nonlinear dynamics and to develop a low complexity algorithm. The project was divided into two segments. First, the orbital and attitude satellite dynamics simulation was carried on Simulink. Then, the new filter block was integrated online in the simulation.

In the context of the European Rover Challenge (ERC) I worked on- development of sensor fusion algorithms and their deployment on ROS platform (on Ubuntu) for mapping Martian terrain with a rover- simulation of Martian terrain through Gazebo and camera sensors through Rviz (on Ubuntu)