Field validation of GNSS-independent positioning enhancement using a wearable ultra-stable quantum magnetometer

Stirling Scholes, Dominic Hunter, Courtney Dyer, Marcin Mrozowski, Allan McWilliam, Phoebe Utting, David Burt, Paul F. Griffin, James McGilligan, Erling Riis, Stuart J. Ingleby · 2026

Read Field validation of GNSS-independent positioning enhancement using a wearable ultra-stable quantum magnetometer by Stirling Scholes, Dominic Hunter, Courtney Dyer, Marcin Mrozowski, Allan McWilliam, Phoebe Utting, David Burt, Paul F. Griffin, James McGilligan, Erling Riis, Stuart J. Ingleby (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-24612.txt.

Introduction

Scholes and colleagues navigate on foot using the Earth's magnetic scars, instead of satellites. Satellite positioning is fragile where signals are jammed or blocked. The crust's permanent magnetic anomalies form a stable, mappable fingerprint, varying from place to place. Their wearable sensor, a free induction decay optically pumped magnetometer built in house, measures the Earth's field precisely and stably enough to use them.

An end-to-end field testing qualifies the sensor. It then measures how much adding its data improves a dead reckoning position estimate, navigation by tracking steps and heading alone. Over a walking route longer than five hundred metres, lasting about six minutes, the fused system achieves a radial position error of only a couple of metres. The error follows the reported spread closely. Scholes's 2026 trial shows a satellite independent position layer, accurate enough to matter when GNSS disappears.

Catalog record

CREATORStirling Scholes, Dominic Hunter, Courtney Dyer, Marcin Mrozowski, Allan McWilliam, Phoebe Utting, David Burt, Paul F. Griffin, James McGilligan, Erling Riis, Stuart J. Ingleby
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.24612

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Field validation of GNSS-independent positioning enhancement using a wearable ultra-stable quantum magnetometer Stirling Scholes, Dominic Hunter, Courtney Dyer, Marcin Mrozowski, Allan McWilliam, Phoebe Utting, David Burt, Paul F. Griffin, James McGilligan, Erling Riis, Stuart J. Ingleby Abstract Increasing the resilience of positioning systems that currently rely on Global Navigation Satellite System (GNSS) signals can be achieved by incorporating stable and sensitive measurements of the permanent crustal anomalies in the Earth’s magnetic field. We have realised this concept using an in-house-developed, wearable, Free-Induction-Decay Optically Pumped Magnetometer (FID-OPM) to carry out precise and stable measurements of the geomagnetic field in a walking trial. We present an end-to-end validation, including qualification of FID-OPM performance, alongside quantification of improvement in accuracy when data from this sensor is added to a dead-reckoning estimation of position. Using our wearable sensor system we achieve a Beckmann-distributed radial positioning error of 2.24 m over a route exceeding 500 m in length and spanning ∼ \sim 360 s. keywords Positioning, Navigation, Quantum sensing, Optically pumped magnetometers 1 Introduction Over the last four decades, Positioning, Navigation and Timing (PNT) applications have become increasingly reliant on Global Navigation Satellite Systems (GNSS). Near-global availability, combined with relatively inexpensive and scalable receiver hardware, has resulted in ubiquitous reliance on GNSS for PNT use-cases ranging over many civil, industrial, and military applications 25 . However, GNSS has a number of vulnerabilities; with a ground-level signal strength typically below -125 dBm, GNSS signals are easily jammed, spoofed or interrupted 16 ; 33 ; 27 ; GNSS coverage is reduced at the Earth’s poles, in built-up areas 17 , and unavailable underground and underwater 41 . Figure 1 : A The Size Weight and Power (SWaP) characteristics of the harness-based wearable system. IMU: Inertial Measurement Unit, RTK GNSS: Real-Time Kinematic GNSS. The inset shows the functional components of the OPM. VCSELs: Vertical Cavity Surface Emitting Lasers, PDs: PhotoDiodes, PBS: Polarising Beam Splitter, ASLs: A-Spherical Lenses, \lambda/4 : Quarter waveplate, B_{0} : Earth field, B_{1} : Enhanced optical pumping field. The scale shows a size comparison to a Consumer Off-The-Shelf (COTS) fluxgate. B The blue boxes labelled i through iii show the processes that lead to the 3D visualisation on the right-hand side. The lower plane shows the magnetic map interpolated from the measured OPM survey (Fig. 3 ). The upper plane shows the surface topography, the ground-truth path (blue), and the magnetic map-matched path (red). The paths start at the green dot and end at the red-square and blue-star respectively. Regions outside of the surveyed area are made transparent. Full analysis of the methodology and results are given in the relevant sections. For use-cases requiring resilient or ubiquitous positioning, GNSS can be augmented by additional positioning aids, which include reference to mapped signals of opportunity, such as magnetic, geodetic or gravitational topography, and reference to fixed ground-based transmitters, for example the enhanced LOng RAnge Navigation (eLORAN) system 34 . Better positional hold-over in the absence of GNSS signals can also be achieved through dead-reckoning, i.e., the integration of platform-mounted inertial sensing and timing references. Across all of these resilient PNT tools, new generations of quantum technologies offer enhanced capabilities; accurate quantum magnetometers 38 ; 22 ; 43 ; 4 and gravimeters 40 offer calibration-free portable sensors, making map matching feasible over extended periods; portable optical clocks 9 ; 30 extend the distribution of accurate phase coherence in beacon networks and underpin improved positioning by dead-reckoning; atom interferometers 23 and nuclear spin gyroscopes 45 augment classical inertial sensor technologies with ultra-low scale factor drift and high bias stabilities. Magnetic map matching, i.e., positioning inference by the comparison of measured time series magnetic data from a moving platform to a fixed map of geomagnetic anomalies, interpolated from prior magnetic survey data, is a useful complement to a suite of resilient positioning tools 5 ; 6 ; 18 and compares favourably to many other alternative positioning aids. Specifically, the crustal lithospheric magnetic anomaly field varies over geological timescales, and is effectively static over the duration of human journeys. Unlike visual positioning, magnetic anomalies are available globally, even over and under the oceans, and do not vary with lighting or weather. Radio and communication signals of opportunity can provide high levels of accuracy, but rely on infrastructure, which can be jammed or degraded. The large spatial extent of geomagnetic crustal anomalies makes them impossible to spoof or degrade (at scale) with anthropogenic magnetic sources. Measurement of these anomalies is effectively passive, and so can be carried out covertly without emission of detectable signals. Lastly, when compared with gravitational anomaly map matching 24 , the contrast of magnetic anomaly signals is significantly higher and can be carried out with higher cadence. Typical crustal magnetic field anomalies are around \sim 100 nT 37 , and can be measured with pocket-sized devices of sensitivity \sim 60 fT/ \sqrt{\text{Hz}} and bandwidths \sim 5 kHz 38 . Crustal gravitational anomalies are around \sim 10 mGal, and can be measured with portable devices having sensitivities of \sim 18 \mu Gal/ \sqrt{\text{Hz}} and bandwidth \sim 1 Hz 36 . Magnetic crustal positioning could therefore be a practical and viable addition to a resilient navigation system, analogous to the manner in which GNSS is currently augmented by independent inertial and timing instrumentation 32 . However, to develop these systems there are a number of technical and integration challenges which must be addressed. The Earth’s magnetic field is subject to variation due to the effects of space weather transients and diurnal fluctuations, and the ionosphere acts as a resonant waveguide for Ultra-Low Frequency (ULF) and Extra-Low Frequency (ELF) magnetic transients 2 ; 3 . For positioning, these are sources of spurious magnetic variation in the pT to nT range. As well as these atmospheric error sources the hardware and systems of the platform may give rise to other unwanted magnetic field variations. Further to these challenges, crustal anomaly reference data, while extant over virtually all of the Earth’s surface 28 ; 1 , consists of survey measurements of widely varying density, survey date, instrumentation, and altitude. Therefore, the use of these reference data requires efficient interpolation methods to allow calculation of expected magnetic anomaly fields over possible positioning locations 21 ; 42 . In the context of these challenges, a successful demonstration of positional aiding using Optically Pumped Magnetometers (OPMs) must demonstrate several parallel capabilities. First, a precise and accurate sensor, integrated with its platform, with effective mitigation of platform-induced magnetic fields. Second, the application of suitable algorithms to interpolate accurate magnetic anomaly survey data and fuse magnetic and inertial sensor data into positional estimates. Third, the quantitative validation of the methodologies used i.e., trials must be conducted with an accurate ground truth and the additional benefit of the OPM integration determined. In this trial we have demonstrated these aforementioned parallel capabilities as shown by the progression of boxes i through iii in Fig. 1 . We report the sensitivity and accuracy of an OPM sensor as qualified in the laboratory against a fluxgate magnetometer of similar form factor and leverage the low Size Weight and Power (SWaP) characteristic of our sensor to develop a wearable system. As shown in Fig. 1 A , the system integrates control and

Read the complete text