Global Navigation Satellite System Antenna Research Articles

Instantaneous Ambiguity Reinitialization and Fast Ambiguity Initialization for L1-L2 GPS Measurements.

This paper presents a PREcise and Fast Method of Ambiguity Reinitialization/Resolution (PREFMAR) for L1 and L2 in GPS measurements. The method determines and ambiguities based on the ambiguity functions: and . These ambiguity functions have been described in detail in this work. The developed method of ambiguity initialization and reinitialization in relative positioning can use Global Positioning System (GPS) measurements from only two satellites and one measurement epoch. To resolve and ambiguities, a variance-covariance (VC) matrix of the float solution is not needed. The size of the search area in the PREFMAR method depends on code and phase accuracy as well as on the GNSS signal frequencies. Therefore, the search area is specific for every double or triple Global Navigation Satellite Systems (GNSS) data frequency. However, this part of the research only presents the ambiguity search area for L1 and L2 of GPS measurements. Additionally, a numerical example has been analyzed in detail with the use of the PREFMAR method and a float solution (, ). Finally, the elaborated algorithms were successfully tested on real L1 and L2 GPS measurements for instantaneous ambiguity reinitialization. The PREFMAR method allows instantaneous ambiguity reinitialization if all satellites lose contact with a GNSS antenna, for short and long baselines. Therefore, the PREFMAR has a great potential for precise real-time kinematic GNSS navigation.

Imprints of COVID-19 Lockdown on GNSS Observations: An Initial Demonstration Using GNSS Interferometric Reflectometry.

The ongoing coronavirus disease 2019 (COVID‐19) pandemic has imposed tight mobility restrictions in urban areas, causing substantial reduction in roadway traffic. Many public parking lots are nearly vacant as people across the world have gone on lockdown since mid‐March. This environmental change may have impacts on Global Navigation Satellite System (GNSS) sensors installed on roof of buildings. Here, we use a monitoring site in Boston to exemplify a likely sensitivity of precise GNSS sensors to their nearby dynamic environments including parked vehicles in parking lots. We show that reduced number of parked vehicles since 23 March has decreased the reflector roughness, resulting in an increase in the reflected signal power whose amplitude is quantified by GNSS interferometric reflectometry technique. The uncertainty of retrieved GNSS antenna height drops with beginning of lockdown, allowing more accurate estimate of reflector height, which could have a general implication for better understanding of the fundamental limitations of the technique.

Nonlinear absolute sea-level patterns in the long-term-trend tide gauges of the West Coast of North America

AbstractThe research issue of which are the present relative and absolute rates of rise and accelerations for North America is here addressed. The data of the 20 long-term-trend (LTT) tide stations of the West Coast of North America with more than 80 years of recorded data are shown. The absolute rates of rise are computed by considering the absolute vertical velocity of Global Navigation Satellite System (GNSS) antennas near the tide gauges, and the relative rate of sea-level rise from the tide gauge signals. The 20 LTT stations along the West Coast of North America show an average relative rate of rise of -0.38 mm/yr., an average acceleration of +0.0012 mm/yr2, and an average absolute rate of rise of +0.73 mm/yr. This is the first paper publishing a comprehensive survey of the absolute sea-level rates of rise along the West Coast of North America using the reliable information of relative sea-level rates of rise from LTT tide gauges plus the absolute subsidence rates from different GNSS antennas close to the tide gauge installations.

Mapping Sea Surface Height Using New Concepts of Kinematic GNSS Instruments

For over 25 years, satellite altimetry observations have provided invaluable information about sea-level variations, from Global Mean Sea-Level to regional meso-scale variability. However, this information remains difficult to extract in coastal areas, where the proximity to land and complex dynamics create complications that are not sufficiently accounted for in current models. Detailed knowledge of local hydrodynamics, as well as reliable sea-surface height measurements, is required to improve and validate altimetry measurements. New kinematic systems based on Global Navigation Satellite Systems (GNSS) have been developed to map the sea surface height in motion. We demonstrate the capacity of two of these systems, designed to measure the height at a centimetric level: (1) A GNSS floating carpet towed by boat (named CalNaGeo); and (2) a combination of GNSS antenna and acoustic altimeter (named Cyclopée) mounted on an unmanned surface vehicle (USV). We show that, at a fixed point, these instruments provide comparable accuracy to the best available tide gauge systems. When moving at up to 7 knots, the instrument velocity does not affect the sea surface height accuracy, and the two instruments agree at a cm-level.

Testing Multi-Frequency Low-Cost GNSS Receivers for Geodetic Monitoring Purposes.

Global Navigation Satellite System (GNSS) technology is widely used for geodetic monitoring purposes. However, in cases where a higher risk of receiver damage is expected, geodetic GNSS receivers may be considered too expensive to be used. As an alternative, low-cost GNSS receivers that are cheap, light, and prove to be of adequate quality over short baselines, are considered. The main goal of this research is to evaluate the positional precision of a multi-frequency low-cost instrument, namely, ZED-F9P with u-blox ANN-MB-00 antenna, and to investigate its potential for displacement detection. We determined the positional precision within static survey, and the displacement detection within dynamic survey. In both cases, two baselines were set, with the same rover point equipped with a low-cost GNSS instrument. The base point of the first baseline was observed with a geodetic GNSS instrument, whereas the second baseline was observed with a low-cost GNSS instrument. The results from static survey for both baselines showed comparable results for horizontal components; the precision was on a level of 2 mm or better. For the height component, the results show a better performance of low-cost instruments. This may be a consequence of unknown antenna calibration parameters for low-cost GNSS antenna, while statistically significant coordinates of rover points were obtained from both baselines. The difference was again more significant in the height component. For the displacement detection, a device was used that imposes controlled movements with sub-millimeter accuracy. Results, obtained on a basis of 30-min sessions, show that low-cost GNSS instruments can detect displacements from 10 mm upwards with a high level of reliability. On the other hand, low-cost instruments performed slightly worse as far as accuracy is concerned.

BDS Space Service Volume characterizations considering side-lobe signals and 3D antenna pattern

Autonomous navigation based on GNSS (Global Navigation Satellite System) is an effective way for spacecraft operation without the assistance of ground facilities. The application of GNSS at high orbit has been defined as GNSS Space Service Volume (SSV). However, the SSV application is extremely constrained by poor satellite visibility. The GNSS side-lobe signals are considered as an effective way to enhance SSV availability. Although side-lobe signals have been utilized unconsciously for the high orbit autonomous navigation, characterizations for the side-lobe signals haven't been comprehensively analyzed. What's more, there is a lack of detailed specifications of GNSS antenna patterns which makes it difficult to have an accurate space mission design. In this paper, a comparative study of BDS-3 (the third generation of the BeiDou navigation satellite system) triple frequency signals with a combination of the main-lobe and the side-lobe for three typical SSV missions is conducted. A high-fidelity 3D BDS antenna pattern is reconstructed, and a comprehensive assessment of the received signal power, Doppler shift, receiver measurement errors, signal availability, etc. are carried out respectively. The simulation results show that the availability of 4 or more satellites and GDOP can be improved significantly as compared to the main-lobe alone case. The receiver measurement errors show that B1C and B3I signals perform better than B2a, and pilot signals are primarily recommended in SSV. The assessment in this paper can serve as the theoretical guideline for both GNSS-based space mission design and onboard receiver design.

Python software tools for GNSS interferometric reflectometry (GNSS-IR)

Global Navigation Satellite System (GNSS) interferometric reflectometry, also known as the GNSS-IR, uses data from geodetic-quality GNSS antennas to extract information about the environment surrounding the antenna. Soil moisture monitoring is one of the most important applications of the GNSS-IR technique. This manuscript presents the main ideas and implementation decisions needed to write the Python code for software tools that transform RINEX format observation and navigation files into an appropriate format for GNSS-IR (which includes the SNR observations and the azimuth and elevation of the satellites) and to determine the reflection height and the adjusted phase and amplitude values of the interferometric wave for each individual satellite track. The main goal of the manuscript is to share the software with the scientific community to introduce new users to the GNSS-IR technique.

Rotating GNSS Antennas: Simultaneous LOS and NLOS Multipath Mitigation

Urban environments present significant challenges to global navigation satellite system (GNSS) positioning. GNSS signals are often obstructed by buildings and other obstacles, leading to reflection and diffraction, which cause major line-of-sight (LOS) and non-line-of-sight (NLOS) multipath positioning errors. We propose a new, precise GNSS positioning technique that can simultaneously mitigate LOS and NLOS multipath errors by rotating the GNSS antenna arm horizontally at a certain angular velocity. The key idea behind this technique is to use the antenna motion to mitigate the LOS multipath effect and detect the NLOS multipath signals. The relative carrier phase between the direct signal and the multipath signal frequently changes when a GNSS antenna is moving. As a result, the effect that the strong reflected signal affects the direct signal can be reduced. We use this characteristic to mitigate LOS multipath errors by generating antenna motion. By rotating a GNSS antenna, the Doppler frequency is produced by the antenna motion. If the received GNSS signal is the NLOS signal, the direction of incoming signals is different from that of direct signals. In this situation, the phase of the generated Doppler frequency will be different, and NLOS signals can be detected by comparison with the ideal Doppler frequency. We conducted an experiment to evaluate the proposed method. The positioning error of the proposed method using LOS multipath mitigation and NLOS exclusion by the rotating antenna was decreased from 18.96 to 2.83 m. In addition, the availability of GNSS positioning increased from 44.29 to 100%.

RTK GNSS-Assisted Terrestrial SfM Photogrammetry without GCP: Application to Coastal Morphodynamics Monitoring

The present article describes a new and efficient method of Real Time Kinematic (RTK) Global Navigation Satellite System (GNSS) assisted terrestrial Structure-from-Motion (SfM) photogrammetry without the need for Ground Control Points (GCPs). The system only requires a simple frame that mechanically connects a RTK GNSS antenna to the camera. The system is low cost, easy to transport, and offers high autonomy. Furthermore, not requiring GCPs enables saving time during the in situ acquisition and during data processing. The method is tested for coastal cliff monitoring, using both a Reflex camera and a Smartphone camera. The quality of the reconstructions is assessed by comparison to a synchronous Terrestrial Laser Scanner (TLS) acquisition. The results are highly satisfying with a mean error of 0.3 cm and a standard deviation of 4.7 cm obtained with the Nikon D800 Reflex camera and, respectively, a mean error of 0.2 cm and a standard deviation of 3.8 cm obtained with the Huawei Y5 Smartphone camera. This method will be particularly interesting when simplicity, portability, and autonomy are desirable. In the future, it would be transposable to participatory science programs, while using an open RTK GNSS network.

Use of GNSS for lunar missions and plans for lunar in-orbit development

In the context of Deep Space Exploration as planned by the International Space Exploration Coordination Group (ISECG), the Moon represents a significant step to feed forward what could be done on Mars. Navigation is a key-technology to enable the cislunar and lunar volume discovery. The use of Earth Global Navigation Satellite System (GNSS) signals at Moon altitude has been extensively studied in the past, within and outside the European Space Agency (ESA). Namely, this interest has been reflected in “The Interoperable Global Navigation Satellite Systems Space Service Volume” booklet issued by the International Committee on GNSS (ICG) in 2018 (United Nations - Office for Outer Space Affairs, 2018). However, the main limitation on accurately performing this kind of analysis has always been the limited knowledge of the actual GNSS transmit antenna patterns outside the main lobe, i.e. side-lobes. ESA published the first patterns of GPS satellites using data collected in GIOVE-A (Unwin et al., 2013) and, more recently, the National Aeronautics and Space Administration (NASA) shared very detailed 3D patterns of all GPS Blocks (Donaldson et al., 2018). This allows both ESA and NASA to perform detailed analysis of the expected visibility of GNSS signals at moon altitude. In particular, NASA published visibility results using a GPS-only receiver in a Moon Transfer Orbit (MTO) demonstrating the potential of GPS-based navigation up to the Moon (considering signals above 22 dB-Hz with a 14dBi antenna) (Ashman, 2018).The present paper presents the results of numerical simulations for a Single-Frequency (SF) receiver in the Deep Space Gateway (DSG) orbit, which is an Earth-Moon L2 Halo orbit, using both Galileo and GPS, for which detailed 3D antenna patterns were used. It demonstrates the importance of considering the azimuthal asymmetry of the GNSS antenna patterns and the necessity of using an interoperable Galileo-GPS receiver at such altitudes. Additionally, a comparative analysis between the frequency bands E1/L1 and E5a/L5 is performed to select the one providing the best results. In conclusion, it shows that a high number of satellites is visible at Moon altitude using a receiver with a 14 dBi antenna and a 15 dB-Hz Carrier-to-Noise density ratio acquisition and tracking threshold. Finally, the second part of the paper provides ESA plans for In-Orbit Demonstration (IOD) of the use of GNSS at Moon altitude, covering both CubeSat missions and the DSG.

Seafloor geodetic network establishment and key technologies

Seafloor geodetic network construction involves the development of geodetic station shelter, network configuration design, location selection and layout, surveying strategy, observation model establishment and optimization, data processing strategy and so on. This paper tries to present main technological problems involved in the seafloor geodetic network construction, and seek the technically feasible solutions. Basic conceptions of developing seafloor geodetic station shelters for shallow sea and deep-sea are described respectively. The overall criteria of seafloor geodetic network construction for submarine navigation and those of network design for crustal motion monitoring are both proposed. In order to enhance application performances of the seafloor geodetic network, the seafloor network configuration should prefer a symmetrical network structure. The sea surface tracking line measurements for determining the seafloor geodetic station position should also adopt an approximately symmetrical configuration, and we recommend circle tracking line observations combined with cross-shaped line (or double cross-shape line) observations for the seafloor positioning mode. As to the offset correction between the Global Navigation Satellite System antenna phase center and the acoustic transducer, it is recommended to combine the calibration through external measurements and model parameter estimation. Besides, it is suggested to correct the sound speed error with a combination of observation value correction and parameterized model correction, and to mainly use the model correction to reduce the influence of acoustic ray error on the seafloor positioning. Following the proposed basic designs, experiments are performed in shallow sea area and deep-sea area respectively. Based on the developed seafloor geodetic shelter and sufficient verification in the shallow sea experiment, a long-term seafloor geodetic station in the deep-sea area of 3000 m depth was established for the first time, and the preliminary positioning result shows that the internal precision of this station is better than 5 cm.

Absolute calibration of timing receiver chains at the nanosecond uncertainty level for GNSS time scales monitoring

The use of global navigation satellite systems (GNSS) for time transfer is widespread, in particular for the computation of the coordinated universal time (UTC). It is an inexpensive, easy-to-implement method to compare distant clocks. This time transfer method can also be applied to the monitoring of the GNSS time scales and of the broadcast UTC conversion parameters. One mission of the French space agency (CNES) is to perform this task for Galileo. For that purpose it is mandatory to carry out an absolute calibration of the propagation delay of a GNSS receiver chain. This propagation delay is the one between the GNSS antenna phase centre and the time reference that feeds the receiver chain. In order to allow a link to UTC, the time reference must be a UTC(k).CNES has developed a method of absolute calibration of GNSS receiver chains, in which the delays of the GNSS antenna, the antenna cable and the GNSS receiver are determined separately, with a total uncertainty in the range of 1 ns (1 – σ). The purpose of this paper is to present the method developed by CNES along with its uncertainty budget for GPS, Galileo and BeiDou signals. A validation in common-clock with actual GPS and Galileo signals is also provided. The absolute calibration is finally applied to an accurate monitoring of the GNSS time scales, the Galileo broadcast UTC conversion parameters and the broadcast Galileo to GPS time offset (GGTO), with associated uncertainties.

A low-cost GNSS buoy platform for measuring coastal sea levels

A low-cost Global Navigation Satellite System (GNSS) buoy platform was developed for measuring coastal sea levels to provide information where logistical, physical and/or financial constraints prevent the application of established tide gauges and satellite altimetry, and where spatially and temporally discrete tidal data are required to support surveys for monitoring coastal morphodynamics. The buoy was constructed from scaffold tubing and four floats, with a central platform for the GNSS receiver and antenna. The overall cost was around £300 (buoy parts £100, GPS/logger £200). The single frequency GNSS receiver was set up to record GPS/GLONASS satellite data at 5Hz. Post-processing was carried out with RTKLIB software to derive solutions for sea level. GNSS buoy performance was evaluated against a reference tide gauge using a Van de Casteele test. The RMSE of 1.4 cm, computed from the differences between the GNSS-buoy and reference tide gauge, demonstrates that this low-cost GNSS buoy tide gauge compares well with previous, more expensive GNSS buoys and is suitable for obtaining local-scale coastal sea level data. In addition to providing coastal data at cm accuracy, the results also indicate the presence of high frequency oscillations in the sea level data caused by coastal geometry and wave-induced disturbance.

Analyzing and overcoming the effects of GNSS error on LiDAR based orchard parameters estimation

Currently, 3D point clouds are obtained via LiDAR (Light Detection and Ranging) sensors to compute vegetation parameters to enhance agricultural operations. However, such a point cloud is intrinsically dependent on the GNSS (global navigation satellite system) antenna used to have absolute positioning of the sensor within the grove. Therefore, the error associated with the GNSS receiver is propagated to the LiDAR readings and, thus, to the crown or orchard parameters. In this work, we first describe the error propagation of GNSS over the laser scan measurements. Second, we present our proposal to overcome this effect based only on the LiDAR readings. Such a proposal uses a scan matching approach to reduce the error associated with the GNSS receiver. To accomplish such purpose, we fuse the information from the scan matching estimations with the GNSS measurements. In the experiments, we statistically analyze the dependence of the grove parameters extracted from the 3D point cloud -specifically crown surface area, crown volume, and crown porosity- to the localization error. We carried out 150 trials with positioning errors ranging from 0.01 meters (ground truth) to 2 meters. When using only GNSS as a localization system, the results showed that errors associated with the estimation of vegetation parameters increased more than 100 % when positioning error was equal or bigger than 1 meter. On the other hand, when our proposal was used as a localization system, the results showed that for the same case of 1 meter, the estimation of orchard parameters improved in 20 % overall. However, in lower positioning errors of the GNSS, the estimation of orchard parameters were improved up to 50% overall. These results suggest that our work could lead to better decisions in agricultural operations, which are based on foliar parameter measurements, without the use of external hardware.

Antenna phase center correction differences from robot and chamber calibrations: the case study LEIAR25

In recent years, the Global Navigation Satellite Systems (GNSS) have been intensively modernized, resulting in the introduction of new carrier frequencies for GPS and GLONASS and the development of new satellite systems such as Galileo and BeiDou (BDS). For this reason, the absolute field antenna calibrations performed so far for the two legacy carrier frequencies, the GPS and GLONASS, seem to be insufficient. Hence, all antennas will require a re-calibration of their phase center variations for the new signals to ensure the highest measurement accuracy. Currently, two absolute calibration methods are used to calibrate GNSS antennas: field calibration using a robot and calibration in an anechoic chamber. Unfortunately, differences in these methodologies also result in a disparity in the obtained antenna phase center corrections (PCC). Therefore, we analyze the differences between individual PCC obtained with these two methods, specifically for the Leica AR-25 antenna model (LEIAR25). In addition, the influence of PCC differences on the GNSS-derived position time series for 19 EUREF Permanent GNSS Network (EPN) stations was also assessed. The results show that the calibration method has a noticeable impact on PCC models. PCC differences determined for the ionosphere-free combination may reach up over 20 mm and can be transferred to the position domain. Further tests concerning the positioning accuracy showed that for horizontal coordinates differences between solutions were mostly below 1 mm, exceeding 2 mm only at two stations for the GLONASS solution. However, the height component differences exceeded 5 mm for four, six and six stations out of 19 for the GPS, GLONASS and Galileo solutions, respectively. These differences are strongly dependent on large L2 calibration differences.

Design of a GNSS Antenna to Prevent LNA Saturation and Intermodulation Caused by S-Band Signals

In this paper, a Global Navigation Satellite System (GNSS) antenna design which prevents low-noise amplifier (LNA) saturation (compression) and intermodulation caused by S-band signals is proposed to improve the reception of GNSS signals. A simulation was conducted to determine which method can more efficiently decrease the S-band signal while maintaining the noise figure and the antenna gain concurrently. Considering the simulation results, an LNA is manufactured and an LNA performance test is conducted. System-level tests are performed to verify the new GNSS antenna function and to assess the performance improvement. The test results showed that the proposed antenna with the new LNA operates robustly without being affected by signals radiated from nearby S-band antennas while also meeting the required specifications.

Modeling of Real Time Kinematics localization error for use in 5G networks

In 5G networks information about localization of a user equipment (UE) can be used not only for emergency calls or location-based services, but also for the network optimization applications, e.g., network management or dynamic spectrum access by using Radio Environment Maps (REM). However, some of these applications require much better localization accuracy than currently available in 4G systems. One promising localization method is Global Navigation Satellite System (GNSS)-based Real-Time Kinematics (RTK). While the signal received from satellites is the same as in traditional GNSS, a new reception method utilizing real-time data from a nearby reference station (e.g., 5G base station) results in cm-level positioning accuracy. The aim of this paper is to obtain a model of the RTK localization error for smartphone-grade GNSS antenna under open-sky conditions, that can be used in 5G network simulators. First, a tutorial-style overview of RTK positioning, and satellite orbits prediction is provided. Next, an RTK localization simulator is implemented utilizing GNSS satellites constellations. Results are investigated statistically to provide a simple, yet accurate RTK localization error framework, which is based on two Gauss-Markov process generators parametrized by visible satellites geometry, UE motion, and UE-satellite distance error variance.

The Effect of Mutual Coupling on the Performance of GNSS Antenna Arrays

Serious mutual coupling effect must be taken into account for the miniaturized adaptive anti-jamming antenna arrays in global navigation satellite system (GNSS) receivers. In this paper, the effect of mutual coupling on the performance of GNSS antenna arrays has been comprehensively discussed. The nulling performance of the GNSS antenna array in presence of mutual coupling is efficiently evaluated by the proposed embedded element pattern method, which is further verified by a simulated miniaturized four-element GNSS antenna array in nonadaptive mode. In the space-time adaptive processing (STAP) array, a generalized wideband signal model based on the dispersion phenomenon is introduced to study the effect of mutual coupling on the optimal STAP array response, signal-to-interference-plus-noise ratio (SINR) loss, and the minimum mean square error (MMSE) of the Wiener filter. Employing the STAP normalized least mean square (NLMS) algorithm, the simulation results show that the mutual coupling leads to a decrease in convergence rate, an increase in misadjustment, and a loss of SINR, whether or not the compensation is considered.

Real-time kinematic positioning algorithm with GNSS and high-frequency accelerometer observations for broadband signals

Real-time kinematic (RTK) positioning technology is widely used for deformation monitoring, but the signals of high-frequency band such as velocity and acceleration are easily polluted due to the bad signal to noise ratio. In order to obtain high-precision and broadband signals for deformation monitoring, we propose an enhanced RTK positioning algorithm by adding high-frequency accelerometer observations and estimate the bias of the acceleration as an unknown parameter in real-time. In addition, a dataset was constructed based on an experiment and used for the validation of our approach. The experiment is operated on a platform and includes a dynamic Global Navigation Satellite System (GNSS) antenna, a low-cost micro-electro-mechanical system (MEMS)-type accelerometer and a slide rail. We conducted eight simulations by sliding the collocated GNSS antenna and MEMS from one point to the other point by manual operation to record 2D movements in horizontal components. A vernier caliper was also set up to estimate the reference displacement for comparison, the maximum sliding distance of the platform is restricted to about 0.5 m. Our validation analyses demonstrate that the enhanced RTK positioning algorithm can provide high-precision displacement, velocity, and acceleration signals in real-time. The average accuracies are 5.0 mm, 0.6 mm s−1 and 1.0 mm s−2 respectively. Furthermore, the high-resolution acceleration signal from the MEMS is helpful to improve the RTK ambiguity resolution because the R-ratio (the ratio of the second minimum to the minimum quadratic form of residuals) values achieved by the new approach are much larger than the standard RTK approach.

Analysis of Tiangong-2 orbit determination and prediction using onboard dual-frequency GNSS data

Tiangong-2 is the first Chinese real space laboratory, launched into earth orbit with an altitude of approximately 393 km on September 15, 2016. Precise orbit determination (POD) is one of the essential conditions for conducting onboard scientific research. The precise orbit of Tiangong-2 is mainly determined by the reduced-dynamic method using space-borne global navigation satellite system (GNSS) data, which are collected by an onboard dual-frequency GNSS receiver. The in-flight performance of the receiver is first assessed in terms of tracking ability, code multipath errors and noise levels of code and carrier phase observations. Then, the effects of refined non-gravitational force models and GNSS antenna phase center variations (PCVs) on the POD are analyzed. The estimated empirical accelerations are substantially reduced by 45% in the along-track direction using a macro-model compared to that using a cannonball model for drag computation. The application of antenna PCVs correction leads to better consistency between reduced-dynamic and kinematic orbit solutions. Satellite laser ranging validation shows that the accuracy of Tiangong-2 precise orbits is 1.7 cm after using the improved reduced-dynamic method. Based on the orbit determination results, the quality of Tiangong-2 orbit predictions is assessed for periods without and with the refined non-gravitational force models. The use of the macro-model and high-quality density models for drag computation is shown to help improve orbit prediction quality.