This record is currently in review state, the data hasn’t been validated yet.
Design, Implementation and Validation of Relative Navigation Algorithms for the CleanSpace One Mission
- Master Thesis
The increasing number of orbital debris poses a real danger to the operation of spacecrafts in low earth orbit. A possible mitigation strategy is the active removal of debris using a spacecraft that is able to find, approach, catch, and de-orbit debris. This thesis investigates the simulation, implementation and optimization of the relative navigation subsystem for on-orbit tracking of debris in the context of the CleanSpace One mission, which is an active debris removal mission with the goal of de-orbiting Swisscube. A simulation framework combining the Robot Operating System with the astrodynamics library Orekit is proposed and implemented. Building on this simulation framework, a state-of-the-art, angles-only relative navigation filter based on the Unscented Kalman Filter and the mean relative orbital elements formulation is investigated and implemented. The angles-only filter is extended by including range measurements, and a novel variant of the needed non-linear transformation between mean and osculating orbital elements based on the Draper semi-analytical satellite theory is proposed. Six different variants of the relative navigation filter are evaluated using randomly generated close- and far-range rendezvous scenarios in the sun-synchronous, low-eccentricity orbit of Swisscube. It could be shown that, while state-of-the-art angles-only tracking is a viable option, the inclusion of range greatly reduces tracking uncertainty and improves the robustness with respect to filter tuning and non-optimal initialization. Additionally, it could be demonstrated that the proposed application of the Draper semi-analytical satellite theory further improves tracking performance by providing a high-precision model of the osculating orbit. Overall, the development and simulation of the relative navigation subsystem for the CleanSpace One mission could be significantly advanced by this work. New insights for the next iteration of mission planning could be obtained by the monte-carlo type evaluation of filter strategies in a wide range of scenarios. Show more
External linksSearch via SFX
SubjectSPACE DEBRIS (ASTRONAUTICAL ENGINEERING); ROBOTICS; orbital mechanics; Kalman Filter
Organisational unit03737 - Siegwart, Roland Y. / Siegwart, Roland Y.
MoreShow all metadata