An Adaptive Aeromagnetic Compensation Method Based on Local Linear Regression

Abstract

Aeromagnetic compensation plays an important role in airborne magnetic survey to eliminate the magnetic interference from the aircraft. However, the aeromagnetic compensation methods at present still cannot suppress the interference to the extremely low noise level of a high-sensitivity scalar magnetometer due to the common assumption about the time-invariation and the linearity of the Tolles-Lawson model depicting the aircraft magnetic interference. In this paper, an adaptive method based on local linear regression is proposed to improve the precision of aeromagnetic compensation. Instead of only using the whole historical calibration data all at once to estimate the model coefficients in advance as the traditional approach does, the proposed method calculates the coefficients in real-time by local linear regression during an aeromagnetic survey, using not only the historical calibration data but also the online measuring data, and both of them are required to be similar in a limited time window. The measured data were used to test the proposed method and the experimental results demonstrated its efficiency.