Magnetic target recognition and localization method unaffected by attitude

Abstract

Abstract This letter presents a magnetic target inversion method that does not vary with changes in the coordinate system and is based on the cross-product of the intermediate eigenvectors of any two points in the dipole field which is in the same/opposite direction as the magnetic moment vector. We used tensor geometric invariants to interpret this new physical property and obtain the unit magnetic moment vector. Using this, the unit vector of the measurement point-source displacement vector was derived. The distance between the measurement point and the source was obtained via the Frobenius norm of the gradient tensor matrix. Simulations verified that the proposed method is unaffected by attitudes and yields unique inversion results, and the results revealed that the inversion accuracy of the proposed method is high. The simulation results also show that conditional cosine and measurement noise have considerable influence on the inversion accuracy in the proposed method.