Reconfigurability in space systems : architecting framework and case studies

Abstract

Reconfigurability in engineered systems is of increasing interest particularly in Aerospace Systems since it allows for resource efficiency, evolvability, and enhanced survivability. Although it is often regarded as a desirable quality for a system, it has traditionally been difficult to quantitatively analyze its effects on various system properties in the early design stage. In order to allow for gaining an in-depth understanding of the various aspects of reconfigurability and its relationship with a system's architecture, a framework encompassing a set of definitions, metrics, and methods has been proposed. Two different modeling schemes, based on Markov models and controls theory, are first developed to show how the states and time aspects of reconfigurable systems can be naturally modeled and studied. An analytical model for quantifying the effect of reconfigurability on mission logistics, specifically spare parts demands, is formulated and it is shown through one specific example that reconfigurable parts can allow for 33-50% mass reduction. The system availability, however, becomes very sensitive to the reliability of the parts. Two case studies are then used for detailed illustration of the application of the developed framework.

(cont.) In the first case study, the effect of reconfigurability on a fleet of planetary surface vehicles for a surface exploration mission are analyzed. It is found that a fleet of reconfigurable vehicles can allow for a mass savings of up to 27% and their expected transport capability degradation is almost three times lower as compared to a fleet of non-reconfigurable vehicles. In the second case-study, the reconfiguration of low earth-orbit communication satellite constellations is considered for evolving to higher capacity levels. It is found that reconfiguring a previously deployed constellation can be a viable option only for certain capacity levels and multi-payload launch capability scenarios. In addition to the high level 'ility' perspectives, a lower level design assessment is also carried out through a survey of 33 representative reconfigurable systems. It is found that on average, for commercial items the cost of reconfigurability is 35%, and the average useful state occupancy time is always at least 10 times the reconfiguration time of the system. Based on the illustrative results of the case studies, and generalization of empirical data, a few principles and guidelines for design for reconfigurability are proposed.