Spacecraft proximity navigation and autonomous assembly based on augmented state estimation: Analysis and experiments

This paper presents a spacecraft relative navigation scheme based on a tracking technique. The augmented state estimation technique is a variable-dimension filtering approach, originally introduced by Bar-Shalom and Birmiwal[1]. In this technique, the state model for a target spacecraft is augmented by introducing, as extra state components, the target's control inputs. The maneuver, modeled as accelerations, is estimated recursively along with the other states associated with position and velocity, while a target maneuvers. By using the proposed navigation method, a chaser spacecraft can estimate the relative position, the attitude and the control inputs of a target spacecraft, flying in its proximity. It is assumed that the chaser spacecraft is equipped with on-board sensors able to measure the relative position and relative attitude of the target spacecraft. The available sensors would provide a measurement update sample time of the order of one second and be subject to random measurement interruption longer than one second. As preliminary analysis, this work introduces the technique applied to the planar, three-degree-of- freedom, spacecraft relative motion. The proposed approach is validated via hardware-in-the-loop experimentation, using four autonomous three-degree-of- freedom robotic spacecraft simulators, floating on a flat floor. The proposed navigation method is proved to be more robust than a standard Kalman Filter estimating relative position and attitude only.