A fast positioning approach based on geomagnetic surface linearization

Most existing geomagnetic navigation algorithms are ineffective when the regional field intensity varies too smoothly in space. To solve this problem, we developed a new algorithm. This algorithm fits a regional geomagnetic surface with linearized planes, then forms linear equation groups to perform the position estimate of a magnetic measurement point. Furthermore, a large-scale accurate positioning precept is proposed, in which a regional geomagnetic map is divided into multi-level submaps to reduce the linearization error. Compared with the existing algorithms, the method has many advantages: higher calculation speed with lower storage capacity request, does not rely on intense anomalous field, and does not need convergence time. Simulations performed with the IGRF model demonstrate the effectiveness of this algorithm.