A hybrid quantum gravimeter kept a ship’s gravity-aided horizontal position within about 1 nautical mile during an approximately six-hour, 45-nautical-mile (83-km) maritime evaluation, according to an arXiv preprint dated 26 August 2026. GNSS, the satellite positioning system, was excluded from the navigation measurement, inertial-navigation, map-matching and tilt-correction chain.
A quantum reference for a moving sensor
The instrument combined an atom interferometer with a classical accelerometer. The atom-based measurement provided atom-referenced stabilization for the accelerometer’s bias, meaning the slow drift in its baseline, while an independent navigation-grade inertial measurement unit supplied the motion mechanization. The same instrument was operated from a 29-m research vessel for navigation, gravity surveying and comparisons between gimbaled and rigid strapdown configurations.
The navigation gap
The contrast was clearest at the endpoint. The unaided INS ended at approximately 14 nautical miles (26 km) of error, while Q-CTRL GravNav ended at 2.2 nautical miles (4.0 km), an approximately 6.3-fold difference in endpoint error. Across the evaluation, the reported gravity-aided horizontal error stayed within approximately 1 nautical mile. These are route-specific results, and no formal uncertainty interval was reported.
The sea survey went finer
The maritime survey used a 280-km coastal route. Four registered profiles came from northbound and southbound traversals, with the instrument run in both gimbaled and strapdown operation. Repeating the route in these configurations gave the researchers a direct comparison of the rigidly mounted and gimbaled measurements.
The maritime quantum-gravity profiles generally followed the grav SWOT 04 satellite map at its resolution. The reported resolved anomalies reached an along-track scale of about 300 metres, roughly 50 times finer than the satellite map’s half-power wavelength. That describes detail measured along the route in these surveys, not a universal resolution for every environment.
The repeatability measurements were reported in milligals, or mGal, the unit used for the gravity readings. The two strapdown traces showed 0.67 mGal in-run stability at a 1,200-second integration time, a 0.16 mGal overall mean difference, and 0.93 mGal stability at 300 seconds. The strapdown traces followed the gimbaled traces without systematic bias or appreciable added high-spatial-frequency variability.
Classical readings and a long stationary check
The classical-sensor-only traces varied significantly and had an overall 570 mGal offset. A separate road survey added a different test: the northbound and southbound profiles differed from an airborne map by 7.7 and 8.2 mGal RMS, respectively, with 6.3 mGal pairwise in-run stability.
An uninterrupted 56-hour stationary laboratory record provided a longer look at drift. The hybrid estimate’s long-term drift was approximately 70 times lower than the classical channel alone. The analysis cautions that stationary laboratory stability may not predict performance while the system is underway.
A field demonstration with clear limits
The no-GNSS result is limited to the navigation chain. GNSS was excluded from the navigation measurement, INS mechanization, map matching and tilt correction. The coastal survey, road survey and stationary record were separate checks, so they should not be read as an extension of that GNSS-free navigation result to surveying.
The results do not establish multi-week fully closed navigation, performance on longer routes with different maps or platform motion, practical size, weight and power for the complete gravimeter, or transfer to autonomous and unmanned platforms. Nor do they establish superiority over classical gravimeters or show that 300-metre anomalies can be recovered universally or necessarily improve navigation.
The authors say, to their knowledge, this is the only publicly reported end-to-end GNSS-independent map-matching navigation result from a mobile quantum gravimeter. The record is an arXiv preprint, version 1, dated 26 August 2026, and its acknowledgments report support from the UK Royal Navy through the Defence and Security Accelerator.
Paper data and sources
Original title: GNSS-free quantum gravity-aided navigation and fine-scale marine surveying with a strapdown quantum gravimeter
Authors: Patrick J. Everitt, Donald H. White, Todd Lyon et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-26
DOI: Not available
Original paper · Full text