Triaxial magnetic electronic compass error compensation method based on depth learning

Abstract

A triaxial magnetic electronic compass error compensation method based on depth learning. An implicit error model is trained for compensating non-linear error in magnetic compass measurement and improving the orientation accuracy of the magnetic compass. The error model training consists of two stages: a first stage is pre-training, and a second stage is reverse trimming by using a back-propagation algorithm for fine-tuning all layers of the network, and reducing the error of model training. The magnetic compass calibration and compensation procedure is to use depth learning algorithm training to obtain a non-linear error model, and the distorted measurement magnetic field is inversed back to a true magnetic field value, thereby reducing the calculation error of course angle. The invention aims at nonlinear error of magnetic compass and provides the error training method based on depth learning; compared to random initialization of a traditional neural network, the weight of each layer locates in a better position of parameter space, so as help to improve the convergence of the algorithm and model training accuracy and achieve high-precision orientation of magnetic compass.

Description

technical field [0001] The invention relates to the error compensation of the magnetic field measurement of the magnetic electronic compass, especially the compensation for the non-linear error part. technical background [0002] High-precision magnetic and electronic compass is not only one of the core technologies of geomagnetic navigation, but also widely used in fields such as mineral exploration, drilling, and robot orientation. Magnetic and electronic compasses with better performance have been developed abroad, but the nonlinear errors caused by soft magnetic interference and cross-axis effects have not been well resolved. Moreover, because the magnetic electronic compass has obvious military potential, Western countries have strictly prohibited high-precision navigation products from being exported to my country. [0003] The magnetic compass uses a three-axis magnetic sensor and an acceleration sensor to measure the components of the geomagnetic field and the gravi...

AI Technical Summary

Problems solved by technology

Although the neural network can identify nonlinear systems and solve complex modeling problems, it generally has problems such as slow convergence and poor real-time performance. Training speed has become the most important problem restricting its development and application.

In addition, in the training process of traditional neural network, there will be "platform" phenomenon or over-fitting, resulting in poor generalization ability of the network.

Images