Magnetometer calibration method based on particle swarm optimization of double-objective function

Abstract

The invention relates to a magnetometer calibration method based on particle swarm optimization of a double-objective function, which belongs to the field of sensor calibration. The method comprises the following steps: analyzing all kinds of errors of a magnetometer and establishing an error model of the magnetometer; establishing a double-objective function in a particle swarm optimization algorithm; and calculating the error parameters of the magnetometer according to a particle swarm optimization algorithm based on the Von Neumann topology. The problem of low accuracy caused by the lack ofrestriction on the three-axis magnetic field component during magnetometer correction by the traditional particle swarm optimization algorithm is solved. By adopting the Von Neumann topology, the situation in which particles can easily fall into local optimum and the parameters cannot be estimated is improved, and the global search ability of the algorithm is enhanced. The method of the inventiondoes not need a large amount of data or good initial value condition and noise distribution, and is easy to operate. The simulation experiments show that the accuracy of parameter calibration can reach more than 95%. It is ensured that the algorithm is simple under large error of a gyroscope. The method has high engineering practicability and a broad application prospect.

Description

technical field [0001] The invention relates to a magnetometer correction method based on double objective function particle swarm optimization, which belongs to the field of sensor calibration. Background technique [0002] The magnetometer can measure the magnetic field components on the three axes of the geomagnetic field according to the inherent geomagnetic field information of the earth, so as to calculate the total geomagnetic field strength and heading angle information, so it is widely used in the attitude reference system. However, in the manufacturing process of the magnetometer, it is difficult to ignore its own errors, and it is extremely susceptible to the influence of the surrounding soft and hard magnets. These factors will affect the measurement accuracy of the magnetometer and the calculation accuracy of the heading angle, which requires compensation and correction before the magnetometer is used. [0003] For the problem of magnetometer compensation and co...

AI Technical Summary

Problems solved by technology

The above-mentioned algorithms have complex operations, many sampling points, and high initial value requirements. For these problems, the literature (Wu Z, Wu Y, Hu X, et al. Calibration of three-axis strapdown magnetometers using particle swarm optimization algorithm[C] / / Robotic and Sensors Environments (ROSE), 2011IEEE International Symposium on.IEEE, 2011:160-165.) proposed to apply the particle swarm optimization algorithm PSO (Particle Swarm Optimization) to the magnetometer calibration. The algorithm selects the appropriate target Function solves the problem of finding the global optimum

Literature (Wu Z, Wu Y, Hu X, etal.Calibration of three-axis magnetometer using stretching particle swarmoptimization algorithm[J].IEEE Transactions on Instrumentation and Measurement,2013,62(2):281-292.) for particles easy to fall into For the local optimal problem, an enhanced particle swarm optimization algorithm SPSO (Stretching Particle Swarm Optimization) is proposed to enhance the robustness of the algorithm, but there are many sampling points and a large amount of calculation

At present, the particle swarm optimization algorithm is used for magnetometer calibration. Most of the objective functions that constrain the total magnetic field are selected, and the constraints on the three-axis magnetic field components are ignored, resulting in the error parameter estimation error.

Images