Named after the Italian explorer Amerigo Vespucci, GPS III-01 is the first out of ten satellites of a new generation of GPS spacecraft. With a body size of 2.5 m x 1.8 m x 3.4 m, a spanwidth of roughly 15 m, and a mass of about 2200 kg, Verspucci is significantly larger and heavier than previous generations of GPS satellites. This is partly attributed to higher signal power as well as a new civil signal on the L1 frequency, namely L1C. Vespucci was launched end of December 2018 and started transmission of navigation signals on 9 January 2019 in the L1, L2, and L5 frequency band. Although the satellite is still set unhealthy, its signals are tracked by a large number of stations of the International GNSS Service (IGS).

Solar radiation pressure (SRP) modeling is a key issue for precise orbit determination. As little is known about detailed surface properties and dimensions, empirical solar radiation pressure models are a viable approach for SRP modeling for this new generation of GPS spacecraft. Therefore, this contribution will analyze the performance of different parameterizations of the empricial CODE orbit model (ECOM-2) in combination with tightly constrained once-per-revolution parameters and compare them with the previous GPS satellites. Furthermore, an emprirically optimized box-wing model will be constructed based on approximate dimensions of the satellite and assumed optical properties of the satellite surfaces.

ESA’s Navigation Support Office (NavSO), located at the European Space Operations Centre in Darmstadt, Germany is one of the leading centres for Precise Orbit Determination (POD). NavSO generates highly precise orbits for all of the Sentinel satellites in support of the European Copernicus program. It is coordinating the Galileo Geodetic Service Provider consortium, responsible for the generation of the Galileo Geodetic Reference Frame and highly precise monitoring products for the Galileo system. Furthermore, it also contributes to the generation of the International Terrestrial Reference Frame (ITRF) via contributions to the International GNSS Service (IGS), the International Laser Ranging Service (ILRS), and the International Doris Service (IDS).

To maintain its role as one of the leading centres for POD, ESA’s Navigation Support Office is permanently improving its orbit prediction and orbit determination capabilities in order to ensure the best possible support for all ESA missions where precise navigation is required and also for the European space programs Galileo and Copernicus.

In this presentation, an overview of NavSO’s latest GNSS POD capabilities will be provided, with a particular attention on challenging scenarios such as small tracking networks (15 – 20 stations), and long orbit arcs. These scenarios have been selected, as in contrast to the POD performed for large networks and short orbit arcs (usually 24 hours), longer arcs will uncover modelling deficiencies. Among others, the presentation will compare the impact of different POD setups, like extended orbit arc lengths, the use of Satellite Laser Ranging observations, the inclusion of Low Earth Orbiting receivers (e.g. Sentinel satellites), impact of improved solar radiation pressure models and also the capability of integer ambiguity resolution.

The Galileo constellation is at the final straight to achieve the official full operational capability with 24 healthy satellites in orbits. Hence, it demands a precise orbit determination strategy which provides the accuracy at the comparable level to that of the operational orbits of GPS and GLONASS systems provided by the International GNSS Service (GNSS). The main challenge in the precise GNSS orbit determination is modeling of the impact of direct solar radiation pressure (SRP) together with albedo and infrared radiation (IR) emitted by the Earth. Additionally, the impact of the navigation antenna thrust has to be considered as it causes a constant acceleration in the radial direction, whereas the misalignment of the normal of solar panels to the Sun direction causes accelerations along two solar panel axes.

We have composed the box-wing model, which considers the direct SRP, Earth albedo and IR, and is based on the Galileo metadata released in late 2017 by the European GNSS Service Centre. The Galileo metadata contains both optical and geometrical properties for the IOV and FOC satellites. Owing to the box-wing model we can consider the actual physical interactions between the solar radiation and the satellite surface elements.

In this contribution, we test the effectiveness of the box-wing model. Additionally, we test what set of the empirical orbit model has to be additionally estimated in order to provide the most sufficient Galileo orbit determination strategy. As a result, we perform a series of the precise orbit determination strategies, i.e., we test the purely analytical model, different types of the hybrid models consisting of box-wing + empirical parameters, and the purely empirical approach employing the Empirical CODE Orbit Model (ECOM2).

The box-wing model can absorb up to 97% of the direct SRP, whereas the residual perturbing accelerations can be absorbed by a small set of the empirically derived parameters which also absorb the misalignment effects of the solar panels. Thanks to the usage of the hybrid solution we significantly reduce the standard deviation of the Satellite Laser Ranging (SLR) residuals to GNSS-derived Galileo orbits for the periods in which the Galileo satellites enter the Earth shadow. The mitigation of the standard deviations of the SLR residuals is at the level of 54% w.r.t the purely empirical model. As a result, using the hybrid solution allows for the Galileo orbit determination with the overall accuracy at the level of 24 mm, as measured by SLR, even for the eclipsing periods.

By the end of 2018, a total of 19 BDS-3 satellites had been launched to complete a preliminary system for global services. And 18 of them are MEO satellites, one is GEO satellite. At present, the multi-constellation Global Navigation Satellite System (GNSS), which generally including GPS, GLONASS, Galileo, and BDS, have more than 100 satellites available. This brings both opportunities and challenges for the Precise Orbit Determination (POD) of GNSSs.
So far there are about 50 stations which including 24 iGMAS stations and 26 MGEX stations can receive the backward-compatible signals from BDS-3 satellites on frequency B1I(1561.098MHz) and B3I(1268.52MHz). What’s more, most of them distributed in the Asia-Pacific Area. Theoretically, the result of GNSSs POD based on either zero-differenced observation or double-differenced observation are same. However, in baseline-creation process, whether the baselines are created taking into account the number of common observations for the BDS3-tracking stations or not has a little effect on GNSSs POD.
With using the BDS3-tracking stations as predefined baselines, the 24-h overlap RMS of BDS3 POD is 28.2, 15.6, and 8.5 cm in along-track, cross-track and radial components, which is better than using no predefined baselines POD by 20%-30%. But with predefined baselines, the 24-h overlap RMS of GPS, GLONASS, and GALILEO is about 5-15% higher than with no predefined baselines. Moreover, with the stations coordinates and tropospheric delay which is estimated by using IGS orbit and clock products at first fixed, the 24-h overlap RMS of BDS-3 POD is 26.2, 13.6, and 7.5 cm in along-track, cross-track and radial components. Meanwhile, with the SLR validation, the mean values are 1.3 and 5.1 cm for GLONASS and GALILEO satellites, respectively, with STD of 4.5 and 8.3 cm. And by using zero-differenced observations, the 24-h overlap RMS of BDS-3 POD is 25.4, 14.8, and 10.5 cm in along-track, cross-track and radial components, and the mean value of SLR validation are 1.5 and 6.2 cm for GLONASSS and GALILEO satellites, and the STD of 5.8 and 9.4 cm, which is slightly better than with double-differenced observations. Considering the limited station distribution, the best POD strategy for GNSSs POD, including BDS-3 satellite, is using the zero-differenced observation.
In this contribution, we present a four-system precise orbit strategy based on observations from iGMAS and IGS MGEX stations. The effects of different processing strategy on the combined precise orbit determination of GNSS satellites are analyzed from the aspects of baselines generation, zero-differenced or double-differenced, etc.