Dynamics, Distributed Control and Autonomous Cluster Operations of Fractionated Spacecraft

Abstract

Fractionated spacecraft deploy satellites' functionalities, such as computation, communication, data storage, payload operations and even power generation, onboard several modules that share those functionalities through a wireless network. With the advent of such an innovative space architecture, non-traditional attributes such as flexibility, robustness and responsiveness, in addition to cost and mass, are introduced to the implementation of space systems, and the equilibrium in the design may shift considerably. In order to enable those non-traditional attributes and thus create huge momentum for fractionated spacecraft, this thesis researches on the autonomous operations of fractionated spacecraft with a focus on cluster reconfiguration. In particular, three aspects have been studied thoroughly to lay the foundation for its implementation. First, functional, physical and organizational architectures of a fractionated infrastructure for long-term Earth observation missions have been proposed, which defines the scenario for our research hereinafter. In the scenario, four fractionated modules are considered with a reference orbit of 800km altitude and the fractionated cluster is regarded as a multi-agent system. Second, the relative motion is studied to provide the knowledge of the modules' long-term flight behavior within the passive cluster. This thesis presents closed-form solutions for the problem of long-term satellite relative motion in the presence of J2 perturbations, and introduces a design methodology for long-term passive distance-bounded relative motion. Last but not least, centralized and distributed approaches to the problem of autonomous cluster reconfiguration are, respectively, developed, both for energy-optimal and time-optimal reconfigurations. In the reconfiguration planning, the non-convex collision avoidance constraints as well as the non-convex final configuration constraints have been taken into account. Theoretical results regarding to the optimality and convergence of developed algorithms have been obtained. All above research areas are devoted to studying, exploiting and enabling the non-traditional attributes of fractionated spacecraft. New results have been contributed to the body of knowledge in all three research aspects. For the development of fractionated space systems, our research will shed light on the cluster design, the autonomous organization of modules within the cluster, and the design of distributed energy- or time-optimal reconfiguration. Even though this thesis is focused on the future-oriented enabling technologies for fractionated spacecraft, the developed methodologies are applicable to other distributed space systems such as formation flying. Therefore, our research can be regarded as a step stone for the implementation of future autonomous distributed space systems.