A Novel Calibration Method for Tri-axial Magnetometers Based on an Expanded Error Model and a Two-step Total Least Square Algorithm

Magnetometer plays an important role in strap-down navigation system, but the measurement results are influenced by various kinds of errors. All combined time-invariance effect inferences are compensated directly by the calibration algorithm, including biases, scale factors, and sensor non-orthogonal errors, as well as soft iron, hard iron, and misalignment errors. This paper proposes a novel magnetometer calibration method based on an expanded magnetometer error model (EM) and a two-step total least square algorithm (TLS) in the magnetic domain. In the method, the EM is derived to achieve calibration directly, and the two-step TLS algorithm is used to calculate the expanded magnetometer EM coefficients. Then, the simulations and experiments are performed to evaluate the proposed method’s performance. The proposed method is compared with the direct least square (DLS) algorithm, which is based on a traditional EM and the least square algorithm (LSA) based on an expanded EM. The results indicate that the proposed algorithm can improve heading accuracy by approximately 13.67% compared with that of the DLS algorithm. Moreover, the proposed algorithm is easy to implement and has resistance to noise interference.