Robust relative navigation for small satellites using line-of-sight measurements

Date

2026-1

Author

Koç, Elif

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

83
views

0
downloads

Cite This

Relative navigation is essential for small-satellite formation flying, rendezvous, and docking. Reliable state estimation must be achieved with limited sensing and nonideal measurements. This thesis investigates robust multiplicative extended Kalman filtering for camera-aided relative navigation using line-of-sight (LOS) unit-vector measurements. A relative dynamics model is fused with LOS observations to estimate relative position, velocity, and attitude, along with gyroscope biases, in a two-spacecraft scenario. The deputy (chaser) carries an onboard monocular camera, and the chief (target) is equipped with active light sources (beacons) that enable LOS-based measurement construction. A multiplicative attitude error representation is used to preserve quaternion consistency and improve numerical stability. Robustness is incorporated to mitigate the effects of faulty LOS observations through adaptive measurement downweighting, and two robustness mechanisms are considered: a single global scale factor and multiple scale factors applied at the beacon level. Simulation studies under nominal and biased-measurement scenarios demonstrate that robust filtering improves estimation accuracy and consistency under directional measurement faults, with the multi-scale formulation providing the greatest benefit when only a subset of beacons is corrupted, while the single-scale formulation is competitive under common-mode corruption. These results support robust close-proximity operations for small satellites.

Subject Keywords

Relative Navigation, Line-of-Sight Measurements, Small Satellites, Robust Multiplicative Extended Kalman Filter, Pose Estimation

URI

https://hdl.handle.net/11511/118778

Collections

Graduate School of Natural and Applied Sciences, Thesis