Design and implementation of small satellite inspection

Abstract

For a variety of missions, vision-based navigation and similar architectures provide the advantage of detailed measurements for a fraction of the size and complexity of ground-based imagers. This thesis provides a simple navigation algorithm using no more than a visual centroid measurement to enable in-situ inspection of space objects. This work evaluates those inspection maneuvers using the Synchronize Position Hold Engage Reorient Experimental Satellites, known as SPHERES. Evaluation of hardware performance was done using data from the International Space Station, in concert with ground-based simulations. Ultimately, this work is in preparation for future experimentation using the VERTIGO vision-navigation payload for SPHERES. The first step presented is an analysis of the measurement capabilities of the SPHERES system and the predicted performance of the VERTIGO system. Using this analysis it is shown that tests run using the former system are applicable to the latter in terms of accuracy, precision, and observability. The second step is an analysis of the tests run on the Space Station, a comparison to those predicted by simulation, and an extension of those results to simulations of more complex maneuvers. Further, a determination of the robustness of the control to disturbances is also performed. Finally, this thesis reflects on the technical and programmatic challenges of developing the VERTIGO payload. From these challenges, lessons are drawn which may guide future developers and program managers, particularly in the university engineering environment.