CEAS EuroGNC 2026
|
Orbit Propagation Strategy for ST3LLARsat1 CubeSat Mission
|
| Josep Peiro Pous |
Aerospace Engineering Department, Universidad Carlos III de Madrid (UC3M), Madrid, Spain. | | Andres Infante Adrian |
Aerospace Engineering Department, Universidad Carlos III de Madrid (UC3M), Madrid, Spain. | | Ghasem Sharifi |
Aerospace Engineering Department, Universidad Carlos III de Madrid (UC3M), Madrid, Spain. | | Behrad Vatankhahghadim |
Aerospace Engineering Department, Universidad Carlos III de Madrid (UC3M), Madrid, Spain. | | Andrés Marcos |
Aerospace Engineering Department, Universidad Carlos III de Madrid (UC3M), Madrid, Spain. |
|
| Abstract:
ST3LLARsat-1 "BOIRA", the first student CubeSat at Universidad Carlos III de Madrid (UC3M), requires a robust onboard orbit determination (OD) capability for its Attitude Determination and Control System (ADCS). This article presents and validates a unified OD architecture centered on a single Simplified General Perturbations #4 (SGP4) propagator, comparing it against an initial dual-propagator (Numerical + SGP4) concept.
The system centers on a Pseudo-TLE Generation Algorithm that converts real-time GNSS data into standardized Two-Line Element (TLE) formats. By using a weighted Levenberg-Marquardt fit, the algorithm ensures that these "pseudo-TLEs" are mathematically precise. This dual-source approach allows the SGP4 propagator to remain highly accurate at all times: using live GNSS data when available, and automatically falling back to traditional ground-based TLEs if the signal is lost.
Using the full Space-Track catalog (31,753 objects), the algorithm reconstructs TLEs with epoch position errors in the [1e-7 ,1e-6] m bin for 99.94 % of cases. In Initial Orbit Determination (IOD) test scenarios using simulated telemetry, a 24-hour data arc allowed for precise drag (B*) estimation. This reduced the 24-hour prediction error to 0.01 km (10 m), whereas a 3-hour arc resulted in a 10.92 km error.
Finally, Model-in-the-Loop (MIL) simulations were conducted to compare both approaches. Under nominal operations, the system received simulated continuous 1 Hz GNSS updates over a 1-hour arc. Under these conditions, both solutions were essentially coincident: the unified SGP4 architecture achieved a Root Mean Square (RMS) algorithmic step error of 5.3275e-6 km (≈ 5 mm) versus 3.4045e-8 km (≈ 0.03 mm) for the numerical propagator. In a 3-day contingency scenario with daily GNSS updates-where actual propagation drift dominates-the SGP4 system proved more robust, exhibiting slower error growth than the numerical integrator.
|
| Keywords: |
 View PDF CEAS-GNC-2026-091 doi: 10.82124/CEAS-GNC-2026-091 |
| Josep Peiro Pous, Andres Infante Adrian, Ghasem Sharifi, Behrad Vatankhahghadim, Andrés Marcos: Orbit Propagation Strategy for ST3LLARsat1 CubeSat Mission. Proceedings of the 2026 CEAS EuroGNC conference. Madrid, Spain. May 2026. doi: 10.82124/CEAS-GNC-2026-091.
|
| BibTeX entry (UTF-8):
@Incollection{CEAS-GNC-2026-091,
author = {Peiro Pous, Josep and Infante Adrian, Andres and Sharifi, Ghasem and Vatankhahghadim, Behrad and Marcos, Andrés},
title = {Orbit Propagation Strategy for ST3LLARsat1 CubeSat Mission},
booktitle = {Proceedings of the 2026 {CEAS EuroGNC} conference},
address = {Madrid, Spain},
month = may,
year = {2026},
doi = {10.82124/CEAS-GNC-2026-091}
}
|