A Sparse Nonlinear Model Predictive Control for Autonomous Space Missions

Propellant consumption minimization is a key factor in space missions, as it strongly affects the duration of any mission. Nowadays, delta-V guidance strategies are obtained by means of classical ground based open loop methods, while academic research has mainly focused on autonomous low-thrust strategies. However, classical methods return instantaneous impulsive thrust actions that are not always feasible in practice, due to the technical limitations of real propulsion systems. In this paper, a novel Nonlinear Model Predictive Control framework for autonomous guidance and control with high-thrust quasi-impulsive maneuvers is presented. The internal prediction model is based on the so-called Modified Equinoctial Orbital Elements, which allow us to overcome relevant singularities given by the standard Keplerian elements. Different NMPC cost functions are compared in order to have a sparse thrust profile, minimize at the same time the propellant consumption and the tracking error with respect to the target orbit. In particular, it is shown how non-quadratic norms could achieve better performances. Finally, an Earth Observation mission, employing different NMPC functionals, is used as a benchmark and the results are compared with the ones coming from the classical astrodynamics solutions.