Definition of optimal LEO constellations for telecom applications using Multiobjective Evolutionary Computation

Satellite constellations are among the most promising space infrastructures because they offer a wide range of services, from Earth observation for scientific purposes to satellite navigation and telecommunications. Services of primary importance for daily life on Earth can be provided by the synergistic operation within the constellation. The following thesis carried out in collaboration with the mission analysis team of SITAEL S.p.A., fits within this engineering panorama. The premises of the work lie in the company's need to equip itself with a tool dedicated to the optimization of satellite constellations in Low Earth Orbit (LEO) for future applications in the telecommunications field. In this regard, the use of Multi-objective Evolutionary Algorithms such as the Nondominated Sorting Genetic Algorithm II (NSGA-II) is proposed to explore sets of constellation designs (Pareto fronts) based on conflicting objectives. Specifically, the focus is on a topological point of view of constellation optimization, leaving the systems engineering point of view to future expansions and improvements of the tool. Therefore, considering a simplified topology of the constellation, made up of reference company satellites equal to each other, the NSGA-II algorithm is used to generate sets of constellations capable of achieving the most global coverage of the Earth, by containing costs and maximizing the operational life of the satellites, which in turn depends on the time spent carrying out mission maneuvers. Case studies in which these conflicting objectives are compared with each other are simulated, showing the evolution of the population, the trend of the nondominated solutions, and the possible configurations of the resulting constellations. In this way, an assessment is carried out of the usefulness of this approach in defining a starting point for future developments of the company tool.