Magnetic sensing on aircraft: getting the engineering right

The platform is the biggest noise source

A magnetometer on an aircraft measures Earth's field plus the aircraft's own: permanent magnetisation, fields induced in ferrous parts, and eddy currents in conductive structure as it manoeuvres.¹ In analysis of magnetic navigation accuracy, calibration error can dominate when sensors are not isolated from electromagnetic interference; one study assumes a conservative 25 nT, or about 5 nT with good sensor placement.²

Compensation

The standard starting point is the linear aeromagnetic compensation model introduced by Tolles and Lawson in 1950, which models permanent, induced and eddy-current terms as functions of the aircraft's attitude and motion.³ Coefficients are fitted from calibration manoeuvres. More recent work extends the model and uses machine learning to capture effects the linear model misses, using open flight data.¹³

Altitude and map resolution

Higher altitude attenuates the fine-scale anomaly features that carry the most positioning information — it acts like a low-pass filter.⁴ Map products differ widely: EMAG2v3 provides a 2-arc-minute global grid referenced to 4 km altitude,⁵ while national surveys can be much finer; Canada's compilation is gridded at 200 m and 1 km.⁶ Mission planning should check anomaly-map resolution and coverage along the intended route.

Integrating with the navigation filter

• Inertial core. Magnetic measurements normally aid an inertial solution rather than replace it.
• Measurement model. The filter needs a model of the map (including its uncertainty) and of residual platform interference.
• Integrity. Map-matching can produce ambiguous or false matches over repetitive terrain; the filter should monitor innovation statistics and reject implausible fixes.
• Temporal corrections. Daily and storm-time field variations should be accounted for from space-weather data or reference stations where available.

Installation and SWaP

• Mount the sensor as far as practical from engines, actuators, power wiring and ferrous structure; survey aircraft often use tail stingers or booms for this reason.
• Route power and data cabling to minimise current loops near the sensor.
• Budget size, weight and power for the complete module and its processing, and account for vibration and thermal range.

Test data you can use today

The US Department of the Air Force–MIT AI Accelerator has published open flight data for magnetic navigation research: twelve flights near Ottawa, Ontario, with optically pumped and fluxgate magnetometers plus INS and GPS truth.⁷ It is a useful baseline for compensation and filter development before flight testing on your own platform.

Limitations

• Compensation quality varies with platform, sensor location and manoeuvre profile; it must be validated on each installation.
• Where anomaly maps are coarse or absent, magnetic aiding provides little benefit.
• Published accuracy figures — such as 13 m DRMS with a high-quality map at low altitude in academic work⁴ — are conditional on those circumstances.

Where Orbital Quantum fits

Orbital Quantum is a Canadian company developing True North Navigation™ quantum magnetometer modules for GPS-independent positioning. The approach reads the structure of Earth's magnetic field with a quantum magnetometer array and matches the live signature against a known field model to resolve position.

Read more on how True North Navigation works, or see Founder's Edition access on orbitalquantum.com.

Sources

1. Gnadt et al. — Signal enhancement for magnetic navigation challenge problem (2020) — arxiv.org
2. Sengupta, P. — A horizontal accuracy metric for magnetic navigation, NAVIGATION 72(4) (2025) — navi.ion.org
3. Gnadt, Wollaber & Nielsen — Derivation and extensions of the Tolles–Lawson model (2022) — arxiv.org
4. Canciani, A. J. — AFIT dissertation (2016) — scholar.afit.edu
5. NOAA NCEI — EMAG2v3 Earth Magnetic Anomaly Grid — www.ncei.noaa.gov
6. Government of Canada Open Data — Canadian aeromagnetic data compilation — open.canada.ca
7. DAF-MIT AI Accelerator — Open flight data for magnetic navigation research (Zenodo) — zenodo.org