A new observation-driven model improves representation of Global Positioning System (GPS) Block IIF satellite yaw during eclipse-season maneuvers, when attitude control departs from nominal yaw-steering law. Using Reverse Kinematic Precise Point Positioning (RKPPP) estimates from 7.5 years of global tracking data, researchers reconstructed yaw time series across 3,935 noon-turn and 4,115 midnight-turn events. They identified recurring mismatches in rotation direction, maneuver onset timing and pre-eclipse yaw evolution, then introduced satellite-specific beta-angle (β-angle) biases and a small-angle midnight-turn decision rule. When tested in network orbit and clock estimation, the revised model reduced attitude-related phase-residual signatures and produced smoother satellite clock estimates around yaw turns. The work supports more consistent precise positioning and timing.
Accurate satellite attitude is essential for Global Navigation Satellite System (GNSS) precise orbit determination (POD), clock estimation, signal-bias modeling and precise point positioning (PPP). Under nominal yaw steering, the navigation antenna points to Earth while solar arrays track the Sun. Near orbit noon and midnight, however, the required yaw rate can exceed hardware limits, and during Earth eclipses Sun sensors may lose a reliable reference. Dedicated noon-, midnight- and shadow-control models therefore take over. If nominal steering is still assumed, carrier-phase wind-up, horizontal phase-center offset corrections and solar radiation pressure can become biased, propagating into satellite clocks and phase biases. Given these challenges, further research is needed on observation-driven yaw models that capture actual Block IIF eclipse-season behavior.
A research team at Wuhan University and collaborating institutions published (DOI: 10.1186/s43020-026-00226-8) the modeling study in Satellite Navigation on 16 October 2026. The team used Reverse Kinematic Precise Point Positioning (RKPPP) yaw estimates from 169 globally distributed International GNSS Service (IGS) tracking stations to assess Global Positioning System (GPS) Block IIF eclipse-season maneuvers. The work targets noon- and midnight-turn discrepancies and tests the revised model in network orbit and clock estimation. The analysis focuses on attitude mismatches that propagate into phase residuals and satellite clock offsets.
The analysis found four recurring discrepancies. Among noon turns, 88 of 3,935 events showed a modeled rotation opposite to RKPPP. Replacing the uniform -0.7° β bias with satellite-specific biases cut mismatches to six, all near switching thresholds. A second issue was an abrupt rotation-direction change at maneuver onset; extending the reversal criterion across the pre-eclipse arc produced a single continuous rotation. Third, 177 events had maximum yaw differences above 40° and 930 between 20° and 40°, mostly at negative β near each satellite’s bias. An empirical piecewise modified β angle, used only before eclipse entry, delayed the modeled maneuver onset and smoothed the yaw profile; afterward all noon events were below 40°, though 95 remained above 20°. For midnight turns, 33 of 4,115 events mismatched when the absolute β angle was near zero. A rule based on the signs of β at maneuver start and β rate resolved 31. In test cases, POD post-fit phase-residual 95th percentiles fell from about 6.6–13.8 cm to 2.9–3.6 cm, and smoothed clock-bias ranges shrank, indicating less attitude mismodeling in orbit and clock products.
The authors said the work turns long-term tracking data into practical model corrections rather than generic assumptions. They said satellite-specific β thresholds matter most near decision boundaries, where small angle differences can flip a maneuver's rotation direction. The revised rules retain the basic maneuver framework while refining noon-turn rotation direction, pre-eclipse yaw evolution, and small-angle midnight-turn direction decisions. They added that gains appear not only in yaw curves but also in reduced POD post-fit carrier-phase residuals and smoother satellite clock estimates. Remaining threshold cases, they said, are targets for refinement as satellite status or onboard control strategies evolve.
For users, the payoff is more consistent GPS attitude information during eclipse seasons, when yaw maneuvers can otherwise leave faint but systematic signatures in precise orbit and clock products. Smoother satellite clocks and lower phase residuals can benefit precise point positioning, timing, signal-bias estimation and carrier-phase wind-up corrections. More broadly, the study highlights the value of observation-driven approaches for assessing and refining GNSS satellite attitude models, especially when manufacturer attitude-control details are limited. The authors caution that the empirical parameters were validated over the analyzed 7.5-year interval and should be reassessed if satellite status or onboard control logic changes. Still, the results support reprocessing efforts and more robust high-precision GNSS products.
###
References
DOI
Original Source URL
https://doi.org/10.1186/s43020-026-00226-8
Funding information
This work is supported by the National Natural Science Foundation of China (No. 42374033), the Major Science and Technology Program of Hubei Province (No. JSCX202501188) and the Natural Science Foundation of Hubei Province (No. 2024AFA023; No. 2025AFA038).
About Satellite Navigation
Satellite Navigation (ISSN: 2662-1363; ISSN: 2662-9291) Satellite Navigation is the official journal of the Aerospace Information Research Institute . The journal aims to report innovative ideas, new results, and progress in the theories, techniques, and applications of satellite navigation. The journal welcomes original articles, reviews and commentaries.
Satellite Navigation
Not applicable
An improved yaw attitude model for GPS Block IIF satellites during eclipse seasons
16-Oct-2026
The authors declare that they have no competing interests.