Hybrid GNSS method preserves centimeter accuracy after correction links drop

Researchers in China say they have built a positioning method that keeps centimeter-level accuracy working for more than 20 minutes after ground correction signals stop. The approach could improve autonomy, surveying, and emergency response in remote areas where reliable communications are not guaranteed.

Why it matters: - High-precision positioning systems can lose centimeter accuracy when correction links fail. - The new hybrid method is designed to keep autonomous vehicles, surveyors, and emergency responders accurately located during communication outages. - The approach could matter most in remote areas, disaster zones, and other places where continuous data transmission is unreliable.

What happened: - Researchers from Southeast University published a study in Satellite Navigation on September 1, 2026. - The study introduces a hybrid positioning framework that combines Undifferenced Network RTK with satellite-broadcast corrections from BeiDou-3 PPP-B2b and Galileo High Accuracy Service. - The method maintained fixed ambiguity solutions for more than 20 minutes after ground-based correction delivery was intentionally interrupted. - The paper reports centimeter-level positioning across 300-kilometer baselines.

The details: - The researchers extended the traditional three-station correction model into a multi-station formulation. - The larger reference-station setup improves atmospheric delay modeling. - Satellite-broadcast orbit and clock corrections remove satellite-related errors before user-side positioning is resolved. - The system maps fixed double-difference ambiguities to undifferenced integers. - That step generates corrections that include fractional-cycle biases, atmospheric terms, and integer ambiguities. - An ionospheric weighted model on the user terminal adapts to correction age. - As corrections get older, the model variance increases so the terminal can separate interpolated ionospheric residuals from age-related drift. - In tests on a 300-kilometer network in western China with five reference stations, the method reached 1.30 cm horizontal accuracy and 5.67 cm vertical accuracy for in-network users. - For an out-of-network station 123 km beyond the nearest baseline, the method delivered 1.014 cm horizontal accuracy and 5.997 cm vertical accuracy. - When corrections were interrupted, the system maintained fixed ambiguity solutions for 1,464 seconds in one test and 1,041 seconds in another. - The broadcast-ephemeris solution lost ambiguity fixing after only a few minutes and later degraded to decimeter-level accuracy. - The paper lists partial support from the National Natural Science Foundation of China, the Fundamental Research Funds for the Central Universities of China, and the SEU Innovation Capability Enhancement Plan for Doctoral Students. - The study DOI is 10.1186/s43020-026-00214-y.

Between the lines: - The work shows a way to blend the stability of satellite-broadcast precise corrections with the local accuracy of network RTK. - That combination could reduce reliance on dense reference-station networks for regional coverage. - It also extends the effective service range of high-precision GNSS beyond the usual 100-kilometer limit of conventional NRTK.

What's next: - The method could be applied in autonomous driving, precision agriculture, marine navigation, and infrastructure monitoring. - Emergency response teams may also benefit if the system proves robust outside controlled test networks. - Wider adoption will depend on how well the framework performs across more regions, more users, and real-world outages.

The bottom line: - The study points to a practical path for keeping centimeter-level GNSS positioning alive after correction signals drop.