Calibration method for a triaxial sensor with selection of a calibration method as a function of the spatial distribution of measurements

Patent metadata

Field	Value
Publication number	US-10838034-B2
Application number	US-201916272322-A
Country	US
Kind code	B2
Filing date	Feb 11, 2019
Priority date	Feb 14, 2018
Publication date	Nov 17, 2020
Grant date	Nov 17, 2020

Abstract

A method of calibrating a triaxial sensor such as a magnetometer. The method includes steps (DIS) to calculate a spatial distribution indicator of a set of measurements output by the sensor, and to determine (AJS), using a fitting method, parameters of a parametric surface that can be used to fit said surface to the set of measurements. The fitting method is selected from among several predetermined fitting methods as a function of the spatial distribution indicator. The predetermined fitting methods may include an ellipsoid fitting method, a sphere fitting method, a sphere fitting method and a two-step fitting method consisting of a first fitting to a sphere followed by a second fitting to an ellipsoid, the centre of which is coincident with the centre of the sphere of the first fitting.

First claim

What is claimed is:

1. A computer-implemented method for calibrating a triaxial sensor, comprising the steps of:
calculating a spatial distribution indicator of a set of measurements output by the triaxial sensor, and
determining, using a fitting method, parameters of a parametric surface that can be used to fit said surface to said set of output measurements,

wherein the method includes a step of selecting the fitting method from among several predetermined fitting methods as a function of said spatial distribution indicator.

2. The method according to claim 1 , wherein the predetermined fitting methods include an ellipsoid fitting method, a sphere fitting method and two-step fitting method consisting of a first fitting to a sphere followed by a second fitting to an ellipsoid the centre of which is coincident with the centre of the sphere of the first fitting.

3. The method according to claim 1 , wherein when the spatial distribution indicator is less than a first threshold, the selected fitting method is the ellipsoid fitting method, when the spatial ellipsoid distribution indicator is more than the first threshold and less than a second threshold, the selected fitting method is the two-step fitting method, and when the spatial distribution indicator is more than the second threshold, the selected fitting method is the sphere fitting method.

4. The method according to claim 1 , wherein calculating the spatial distribution indicator comprises determining a correlation matrix of measurements of said set and determining the condition number of the correlation matrix, said condition number forming the spatial distribution indicator.

5. The method according to claim 1 , further including a preliminary step (SLT) of selecting the measurements of said set among the measurements output by the triaxial sensor, said preliminary step comprising estimating a change in the orientation of the triaxial sensor between the output measurements and selecting measurements associated with sensor orientations deemed to be sufficiently different by comparison with a threshold.

6. The method according to claim 5 , wherein the triaxial sensor is a magnetometer and estimating a change in the sensor orientation between the output measurements includes determining a difference between the amplitude of an output measurement component and the amplitude of a component of at least one of previously selected measurements.

7. The method according to claim 5 , wherein the triaxial sensor is a magnetometer and estimating a change in the orientation of the magnetometer between the output measurements makes use of measurements output by a gyrometer, or uses both measurements output by the magnetometer and measurements output by an accelerometer.

8. The method according to claim 1 , further including determining of a test transformation that transforms the parametric surface fitted to said set of output measurements into a unit sphere, applying the test transformation to the measurements of said set to obtain transformed measurements and determining a degree of fit of the transformed measurements to the unit sphere.

9. The method according to claim 8 , further including applying a current calibration transformation to the measurements of said set to obtain calibrated measurements, determining a degree of fit of the calibrated measurements to the unit sphere and selecting the test transformation as a new current calibration transformation if the degree of fit of the transformed measurements to the unit sphere is more than the degree of fit of the calibrated measurements to the unit sphere.

10. A computer program including instructions for execution of the method according to claim 1 when said program is executed on a computer.

11. A device comprising a data processing unit configured to use the method according to claim 1 .

Assignees

• Commissariat Energie Atomique

Inventors

• Leirens Sylvain

Classifications

• G01C17/38 Primary

  Testing, calibrating, or compensating of compasses · CPC title

• G01R33/0005

  Geometrical arrangement of magnetic sensor elements; Apparatus combining different magnetic sensor types (G01R33/0206 takes precedence) · CPC title

• G01P21/00

  Testing or calibrating of apparatus or devices covered by the preceding groups · CPC title

• G01R33/0206

  Three-component magnetometers · CPC title

• G01C25/00

  Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass (testing, calibrating or compensating compasses G01C17/38) · CPC title