Toward the 1-cm Galileo orbits: challenges in modeling of perturbing forces

Abstract

Precise orbit determination demands knowledge of perturbing forces acting on the satellites of the Global Navigation Satellite Systems (GNSS). The metadata published by the European GNSS Agency for the Galileo satellites allow for the composition of the analytical box-wing model dedicated for coping with the direct solar radiation pressure (SRP), albedo, and infrared radiation (IR). Based on the box-wing model, we evaluated both the magnitude and the characteristic periods of accelerations caused by all the aforementioned forces. We assess which perturbations can be absorbed by the extended Empirical CODE Orbit Model (ECOM2) and what are the consequences of neglecting higher-order ECOM2 coefficients. In order to evaluate the impact of SRP, albedo, IR, and the navigation antenna thrust, we perform a series of precise Galileo orbit determination strategies for Galileo In-Orbit-Validation (IOV), Full Operational Capability (FOC), and two FOC satellites launched into eccentric orbits. The proposed box-wing model is capable of absorbing approximately 97% of the SRP in the Sun-satellite direction, whereas the rest can be mitigated by an additionally estimated small set of empirical parameters. The purely physical box-wing model does not fully handle satellite misorientation and re-radiation effects, such as Y -bias, solar panel rotation lag, that is the misalignment causing a constant acceleration perpendicular to the solar panel axis and the direction to the Sun. However, the box-wing model is especially crucial in terms of the absorption of the higher-order terms of SRP and stabilizes the orbit solutions during the eclipsing periods. Based on the SLR residual analysis, we found a systematic effect at the level up to 50 mm resulting from the omission of the high-order empirical orbit coefficients. We also found that the impact of the albedo, IR, and transmitter antenna thrust on the Galileo orbits reach the level of 5, 14, and 20 mm, respectively. Eventually, we obtain the overall accuracy of the Galileo-FOC orbits at the level of 22.5 mm, even for the eclipsing period for the solution which considers the box-wing model with the estimation of the constant empirical accelerations.

Geocenter coordinates derived from multi-GNSS: a look into the role of solar radiation pressure modeling

“…Table 1 summarizes the naming of test cases, which differ in the approaches for the SRP modeling. We chose the different variations of the ECOM2 model (Arnold et al 2015), which is employed by most of the analysis centers of the International GNSS Service (IGS; Johnston et al 2017) for operational GNSS products, as well as the hybrid approach for SRP modeling as defined by Bury et al (2020). The ECOM model decomposes the accelerations acting on satellites in the DYB frame, i.e., D-pointing toward the sun, Y-along the solar panel rotation axis, and B-perpendicular to D-and Y-axes, completing the right-handed orthogonal frame.…”

“…The most recent and coherent information about the new GPS and GLONASS generations is provided in the frame of the IGS repro3 activities (http://acc.igs.org/repro 3/PROPB OXW.f). By the example of Galileo, the box-wing model accounts for approximately 97% of the direct solar radiation in the D direction (D 0 ) and also for a large part of the periodic terms D 2C and B C (Bury et al 2020). However, the box-wing alone is insufficient without additional correction model parameters.…”

System-specific systematic errors in earth rotation parameters derived from GPS, GLONASS, and Galileo

“…The amplitude of the 3 cpy signal is higher by a factor of 1.9 in the series of the GLONASS-based X pole coordinate when a 3-day arc solution is applied instead of 1-day. These issues may constitute a side effect of the deficiencies in orbit modeling for GLONASS and Galileo (Bury et al 2020;Dach et al 2019). If some significant orbit mismodeling exists, its effect might be amplified when using long arcs because the same number of parameters has to absorb errors accumulated over a long period.…”

“…The composed box-wing model considers the direct SRP, earth albedo, and infrared radiation, based on the Galileo metadata released in the years 2017-2019 by the European GNSS Service Centre. The hybrid strategy of combining both specific terms of empirical and physical models is beneficial for the orbit quality (Bury et al 2020;Li et al 2019). The selection of the set of empirical parameters, which should be chosen in the hybrid solution, is extensively studied by Bury et al (2019).…”

“…In the 3-day GLO and GAL/ GAB solutions, we also see some increased spurious signals for the periods in the range of 20-30 days. The orbit modeling deficiencies might cause all of these issues, i.e., SRP mismodeling for GLONASS and Galileo (Bury et al 2020;Prange et al 2017). If some significant orbit mismodeling issues exist, their effect might be amplified with long arcs because the same number of parameters must absorb errors accumulated over a longer period.…”