Long-distance geomagnetic navigation: Imitations of animal migration based on a new assumption

Plenty of evidence has implied that animals like pigeons and sea turtles utilize geomagnetic information for long-distance migration and homing. In this paper, we conduct an ab initio investigation to imitate a migratory animal's geomagnetic homing across a long distance over the surface of the Earth. Based on a simple assumption that such animals have a natural ability to sense and compare the included angle between the vector geomagnetic field (GF) and a specific geographic direction such as the due north, we show that a long-distance geomagnetic navigation (GN) can be accomplished without any assistance of prestored geomagnetic and geographic information, as long as the navigation is inside a vast area where such a spatial angle can uniquely determine a geographic location. Without any requirement of measuring the strength of the GF, this GN will not be visibly affected by the recurrent fluctuation of the GF that severely hampers traditional GN. By introducing the extended-Kalman-filter algorithm, it is able to find a shortcut in the geomagnetic space, overcoming interferences of strong Gaussian and local-area geomagnetic anomaly. We believe that the proposed approach will find a wide range of potential applications in global, local, and indoor navigations based on magnetic field in the future.