Genetic-Algorithm-Based design for rideshare and heterogeneous constellations

Abstract

Responsive and resilient space-based systems are needed to satisfy changing mission requirements and react to unforeseen challenges. Nontraditional constellations, such as heterogeneous and rideshare constellations, have the potential to increase resiliency and responsiveness, as can using commercial off-the-shelf parts. However, these constellation types introduce design challenges that cannot be solved by analytical techniques. A novel methodology for optimizing these constellations using a multi-objective genetic algorithm and model-based systems engineering techniques is introduced. Included in this methodology are new techniques for measuring constellation resiliency and rapidly determining reachability between two orbits. The methodology is applied to a validation case of designing a Mars-orbiting position, navigation, timing, and communications constellation. The methodology is then used to examine the viability of using rideshare opportunities to construct a satellite constellation. The constellation must provide regional imaging coverage in the event of a natural disaster, and its efficacy is evaluated both in the nominal case and in the case when multiple satellites have failed.