OPTIMIZATION OF THE ARCHITECTURE OF SATELLITE CONSTELLATIONS OF SPACE MISSIONS

Abstract

The deployment of multi-satellite constellations in Low Earth Orbits (LEO) is a pivotal trend in the modern aerospace industry, driven by the advancement of global internet systems, real-time monitoring, and the concept of Space Edge Computing. The efficiency of such missions is intrinsically linked to the spatial configuration of the constellation, as subjective or random selection of orbital parameters leads to inefficient coverage overlaps and increased data latency. This paper addresses the scientific and practical problem of minimizing the time required to achieve full cumulative coverage of a target Earth surface area. The object of the study is the process of spatial maneuvering and the geometric interaction of satellite coverage zones. The study employs mathematical modeling of orbital mechanics (accounting for J2 precession) and stochastic optimization methods. To find the global extremum of the objective function, two approaches—Simulated Annealing and Genetic Algorithm—were implemented and compared. The selection of GA parameters, including population size, mutation rate, and elitism, is justified to balance computational complexity and solution accuracy. A software suite featuring Mercator projection and 3D environment visualization has been developed. It is experimentally proven that the proposed algorithms reduce the time to reach target coverage by 30–40% compared to regular grid-based structures. The study established that the Genetic Algorithm provides 5–10% higher optimization accuracy, while Simulated Annealing demonstrates faster convergence, which is critical for real-time mission planning systems. The findings can be implemented in the design of satellite communication systems, environmental monitoring networks, and infrastructure deployment for orbital AI servers.