Robotics and On-Board Autonomy: For What and How Far Can We Go?

Abstract

The paper addresses the ongoing transition from ground based to robotic on- board autonomous control of space missions. It describes expectations and refers to experiences made with ESA projects. Subsequent parts deal with possibilities and case examples for future on- board autonomy expansions i.e. for going to a higher degree of robotic control. Such new applications however, demand for the mastering of a series of technical challenges which are individually addressed in the paper. Overall robotic on- board autonomous control of space missions offers, as an enabling technology new space mission possibilities and contributes significantly to keep ground segment and operation cost low in the domains where several/ many identical satellites are utilised simultaneously (e. g. LEO constellations). For tailor made single satellite missions however, the balance between ground segment cost reduction and additional autonomy development cost appears not to be attractive. Thus only autonomy functions needed for performance reasons should be implemented for such projects. This applies as long as there is no portability of on- board software from one project to another one. Furthermore the fact that complex on- board autonomous control processes might become major risk sources causing a lowering of the overall mission success probability deserves also special attention.