Performance compensation of GMR-based magnetic azimuth measurement system

Abstract

The magnetic azimuth is one of the important parameters of the directional navigation. This paper proposes a magnetic azimuth measurement system based on the GMR sensor. With the thoughts of information fusion, we study the performance compensation method of magnetic azimuth measurement system based on the radial basis function (RBF) neural network and the BP neural network, and then establish a coupling disturbance compensation model of the magnetic field and the temperature. The experimental results illustrate that the maximum full-scale error of sensor output without compensation is ±21.3%, and the maximum full-scale error after the coupling compensation of the BP neural network and the RBF neural network are ±2.72% and ±0.52% respectively.