Calibration of magnetic compass using an improved extreme learning machine based on reverse tuning

Abstract

Purpose The sources of magnetic sensors errors are numerous, such as currents around, soft magnetic and hard magnetic materials and so on. The traditional methods mainly use explicit error models, and it is difficult to include all interference factors. This paper aims to present an implicit error model and studies its high-precision training method. Design/methodology/approach A multi-level extreme learning machine based on reverse tuning (MR-ELM) is presented to compensate for magnetic compass measurement errors by increasing the depth of the network. To ensure the real-time performance of the algorithm, the network structure is fixed to two ELM levels, and the maximum number of levels and neurons will not be continuously increased. The parameters of MR-ELM are further modified by reverse tuning to ensure network accuracy. Because the parameters of the network have been basically determined by least squares, the number of iterations is far less than that in the traditional BP neural network, and the real-time can still be guaranteed. Findings The results show that the training time of the MR-ELM is 19.65 s, which is about four times that of the fixed extreme learning algorithm, but training accuracy and generalization performance of the error model are better. The heading error is reduced from the pre-compensation ±2.5° to ±0.125°, and the root mean square error is 0.055°, which is about 0.46 times that of the fixed extreme learning algorithm. Originality/value MR-ELM is presented to compensate for magnetic compass measurement errors by increasing the depth of the network. In this case, the multi-level ELM network parameters are further modified by reverse tuning to ensure network accuracy. Because the parameters of the network have been basically determined by least squares, the number of iterations is far less than that in the traditional BP neural network, and the real-time training can still be guaranteed. The revised manuscript improved the ELM algorithm itself (referred to as MR-ELM) and bring new ideas to the peers in the magnetic compass error compensation field.

Short-Term Load Forecasting of Distributed Energy System Based on Kernel Principal Component Analysis and KELM Optimized by Fireworks Algorithm

“…In the kernel KELM algorithm, it is not necessary to give the specific form of the feature mapping function h(x) of the hidden layer node, and the output value can be gotten only by knowing the specific form of the kernel function [35]. Furthermore, so the kernel function is directly in the form of inner product, it is not necessary to set the number of hidden layer nodes, nor to set the initial weight and bias of hidden layer.…”

Two-step calibration method for three-axis magnetic sensor error based on particle swarm optimization

“…Because there are only 9 independent coefficients in the ellipsoidal equation, it is impossible to completely determine the 12 unknown parameters in the error model. The nine coefficients of the ellipsoidal equation can be obtained by the optimal estimation, but the error matrix obtained by different matrix decomposition methods is not unique and a rotation matrix is needed between each other (Liu et al , 2019). Feng et al (2013) used a constrained least square method to fit the ellipsoid, but took the calibration matrix approximately as a lower triangular matrix to get nine error parameters.…”