Partide swarm optimization algorithm in calibration of MEMS-based low-cost magnetometer

In land platform navigation, many systems as Global Positioning System (GPS) and Inertial Navigation Systems (INS) are used to get the position and the orientation solutions. Magnetometers are complementary sensors used in the navigation algorithms to achieve heading information based on utilizing Attitude and Heading Reference System (AHRS) model. However, low-cost magnetometers decrease the precision of the navigation system due to their inherent errors. Thus, a calibration process should be conducted as a first step to compensate the deterministic errors. This paper proposes a method to calibrate a three-axis magnetometer using the Particle Swarm Optimization (PSO) algorithm and the International Geomagnetic Reference Field (IGRF) model. The improved PSO represents the main contribution of the proposed method, which allows the determination of the calibration parameters for each magnetometer data, and the IGRF model is used to determine the true total Earth's Magnetic Field (EMF) in each time step. In this work, we compare the precision of the standard PSO against the proposed method which provides higher robustness by achieving a better compensation of the errors effects (hard and soft iron, etc.). Since the hard and soft iron are the most significant errors in a Micro-Electro-Mechanical Systems (MEMS) based low-cost magnetometers, the proposed method aims to compensate these errors with a minimal error relative to the reference EMF. Several tests have been made to evaluate the performance of the proposed method. The raw measurements of a MEMS-based on the low-cost magnetometer have been collected in Montreal (Canada) using a real car in different environments (under high-voltage lines, city center, highway, tunnel, etc.) full of distortion sources for the magnetic field. The proposed method always gets a better accuracy and precision even in a harsh environment for a low-cost magnetometer.