On-line calibration algorithm, medium and system for magnetometer based on unscented Kalman filter

Abstract

The invention requests to protect a magnetometer on-line calibration algorithm based on unscented Kalman filtering, and the algorithm comprises the following steps: in an initial stage, randomly rotating a magnetometer in a space, and collecting data for a period of time; carrying out ellipsoid fitting according to the collected magnetometer data; an error parameter matrix is obtained according to the magnetometer error model and the fitting result, and initial calibration of the magnetometer is completed; estimating errors in real time by using unscented Kalman filtering in the advancing process, and calibrating the magnetometer; and estimating the course by using the calibrated output of the magnetometer, and correcting the course calculated by the gyroscope. Experimental results show that compared with a pedestrian navigation algorithm in which the magnetometer is not calibrated and only initially calibrated, the magnetometer calibration method provided by the invention has the advantages that closed-loop errors are respectively reduced by 6.17% and 2.8%, the magnetometer is effectively calibrated, course angle divergence in pedestrian navigation is inhibited, and navigation precision is improved.

Description

technical field [0001] The invention belongs to the technical field of inertial navigation and positioning, in particular to an online calibration algorithm of a magnetometer based on an unscented Kalman filter. Background technique [0002] Inertial navigation and positioning technology is an autonomous navigation and positioning technology that does not depend on external information. It completes position calculation by collecting angular velocity and acceleration information of the human body. However, due to the sensor error and the calculation principle, the heading angle error will accumulate over time, resulting in a decrease in positioning accuracy. [0003] The magnetometer can determine the heading by measuring the strength and direction of the magnetic field. There is no cumulative error problem, and it can be used to correct the heading angle. However, due to the error of the magnetometer, it needs to be calibrated before use. [0004] For the calibration of t...

AI Technical Summary

Problems solved by technology

Most of the above algorithms are verified in the laboratory environment, or with the aid of a turntable, which cannot eliminate the magnetic error caused by the dynamic environment during the pedestrian walking process, so it is not conducive to practical use

[0005] The patent publication number is CN 109781084 A, "A Device and Method for Calibrating a Magnetometer". The magnetometer circuit board is installed on the motor, and the magnetometer is calibrated through the rotation of the motor, and the calibration device can also be trigg

Images