Localization Exploiting Spatial Variations in the Magnetic Field: Principles and Challenges

Abstract

Signal processing has played, and continues to play, a fundamental role in the evolution of modern localization technologies. Localization using spatial variations in the Earth's magnetic field is no exception. It relies on signal-processing methods for statistical state inference, magnetic-field modeling, and sensor calibration. Contemporary localization techniques based on spatial variations in the magnetic field can provide decimeter-level indoor localization accuracy and outdoor localization accuracy on par with strategic-grade inertial navigation systems. This article provides a broad, high-level overview of current signal-processing principles and open research challenges in localization using spatial variations in the Earth's magnetic field. The aim is to provide the reader with an understanding of the similarities and differences among existing key technologies from a statistical signal-processing perspective. To that end, existing key technologies will be presented within a common parametric signal-model framework compatible with well-established statistical inference methods.

Figures (5)

• Figure 1: Examples of the spatial variations in the magnetic field-magnitude indoors and outdoors. Also shown in (\ref{['fig:visionen_field']}) are the three field components along the trajectory indicated by the black arrow. Also shown in (\ref{['fig:granso_field']}) is an example of temporal variations in the magnetic field during 30 minutes. Note the significant difference in length and magnitude scale indoors and outdoors.
• Figure 2: Illustration of localization using magnetic-field map-matching. The inference algorithm alternates between comparing the magnetometer measurement $y_t$ to the map $\mathcal{M}$ and predicting the next state $x_{t+1}$ using the motion dynamics $p(x_{t+1}\mid x_t,u_t)$. The graph below each image shows the dead reckoning (edges) and measurement (nodes) information available at that point.
• Figure 3: At top, an array with 30 vector magnetometers measuring the magnetic field at two consecutive locations. Below, a conceptual illustration in 1D of how the displacement $\delta r_t$ can be inferred by fitting a local magnetic field model $\mathcal{M}\xspace_t$ to the measurement from the two locations, and treating the displacement $\delta r_t$ as an unknown parameter in the model.
• Figure :
• Figure :