Research on aeromagnetic compensation of a multi-rotor UAV based on robust principal component analysis

Abstract

Unmanned aerial vehicle (UAV) aeromagnetic survey has been widely used in the fields of mineral resources survey, engineering investigation, and military target detection due to its advantages of high safety, low cost, and wide coverage. However, the UAV carrier can cause severe magnetic interference to a magnetometer, and the interference error has to be removed through aeromagnetic compensation. Since there is complex collinearity in the common Tolles–Lawson aeromagnetic compensation model, the traditional principal component analysis method can effectively eliminate the correlation between variables and improve the compensation effect, but the algorithm is sensitive to noise, and data robustness is unstable. To overcome this problem, this paper proposes a method based on robust principal component analysis to improve the accuracy and robustness of aeromagnetic compensation. The effectiveness and superiority of the proposed method are verified by tests on measured data.