CORRECTION OF AXIAL DISPLACEMENT ERRORS OF A DIGITAL MAGNETOMETER USING LABVIEW

Abstract

When using a three-axis MEMS (Micro-Electro Mechanical Systems), digital magnetometer, in a working electro- magnetic environment, it is necessary to calibrate this device. In our case, the GY-273 module on the QMC5883L chip is used. For this, you can use several programs that allow you to calculate correction factors that take into account two main types of interference – hard-iron and soft-iron distortions. The most common method of calibrating a magnetometer is ellipsoid fitting – rotating the sensor in all directions to obtain a sufficient amount of data that is placed on the surface of the spheroid. Correction factors allow you to transform the spheroid into a regular sphere with the center at the origin.But using these calibration methods, for example, using the Magneto program, does not always give the desired result.The measurement error can remain quite significant and requires additional correction. When performing calibration measurements under conditions of external magnetic influences, situations arise when the coordinate axes in the space of measurement data are not mutually perpendicular and do not have a common intersection point. At the same time, the measured values of the components coming from the sensor are separate vectors that have their own deviations from the axes, both relative to the origin of coordinates and relative to the angles of inclination. In this work, dependencies are proposed for calculating correction matrices that take into account the features of the location of the measuring axes in space. Unlike existing correction methods, the shifts of the centers of the segments of each axis along each coordinate are taken into account, i.e. the shift matrix has a dimension of 3 × 3. This allows you to influence the measurement error. A set of virtual devices has also been developed for visualizing the calibration process, obtaining correction matrices based on the obtained analytical dependencies and processing magnetometer data in real time according to the proposed algorithm. The developed complex can be used in the educational process for laboratory practice for calibrating three-axis sensors and visualizing the three-dimensional measuring space.