Improvement and Omnidirectional Analysis of Magnetic Gradient Tensor Invariants Method

Abstract

The scalar triangulation and ranging (STAR) method is a magnetic detection method based on magnetic gradient tensor invariants, which has a wide application prospect. Due to the asphericity parameter, there are distance error and direction error in the STAR method. There have been studies to compensate the direction error, but the distance error has not been compensated. This article proposes a new STAR method (NSM) that uses an iterative method to compensate both the direction error and the distance error. Considering the convenience and feasibility of the operation, previous researchers analyzed the detection accuracy of the magnetic detection method by the detection error of the magnetic target on a latitude trajectory (latitude line in the spherical coordinate system of magnetic detection). But different latitude trajectories will lead to different analysis results. In order to analyze the influence of motion trajectory on the detection error, the omnidirectional magnetic detection model (OMDM) is established. The omnidirectional error expectation is used to more accurately measure the detection accuracy of the magnetic detection method. Based on the OMDM, the warp trajectory (warp line in the spherical coordinate system of magnetic detection) that accurately measures the detection accuracy is found. Simulation results show that NSM reduces the localization error of the STAR method, Lv-STAR method, and Wang-STAR method by 87.5%, 58.8%, and 12.5%, respectively. Both simulation and experimental results show that when comparing and analyzing the detection accuracy of the magnetic detection method, the results obtained by a latitude trajectory are one-sided, but a warp trajectory is both operable and accurate.