It is of great significance to measure real-time surface vessel magnetic field characteristics based on underwater vehicle equipped with magnetometers. In order to obtain the position information of the vehicle at the bottom of the vessel, it is crucial to estimate its speed and apply it to the auxiliary navigation of acoustic positioning systems. In this paper, we proposed a novel method for estimating moving velocity of an underwater vehicle by processing anomaly magnetic field signals measured by two magnetometers. Firstly, we established a detecting model for anomaly magnetic field of a fixed magnetic object using two magnetometers at different positions on the same straight line. Theoretical analysis based on the detecting model indicated that the speed of underwater vehicle can be obtained by fusing the spatial distance of the magnetometers and the time interval of the anomaly magnetic field waveforms. Then, the sparrow search algorithm was introduced to estimate the time interval considering that this optimization algorithm has the advantage of simple structure and fast convergence. Finally, computer simulation and real-word experiments have been conducted to verify the performance of the method. The results demonstrated that the method can calculate speed of the magnetometers correctly with an error of less than 2% and calculation time less than 3s. The features of easy implementation and low computational complexity make the proposed method a potential candidate for the application in the underwater assisted navigation and other fields.

In order to solve the problem that the cesium optical pump magnetometer is disturbed by the carrier’s interference magnetic field during magnetic field anomaly detection, an interference magnetic field compensation method based on an improved mayfly optimization algorithm (IMOA) was proposed in this paper. First, by combining the measurement results of the attitude sensor with the geomagnetic inclination and magnetic declination in the locality, the measurement results of the optical pump magnetometer can be decomposed into the component values under the three axes of the carrier coordinate system. A compensation model including the carrier interference magnetic field was established. Then, considering the poor global search performance that existed in the mayfly optimization algorithm (MOA), an elite chaotic reverse learning strategy and Levy mutation strategy were introduced to improve the MOA. The compensation performance of the IMOA was estimated with a series of field experiments and compared with the stretching particle swarm optimization algorithm. The experiment results indicated that these two methods can effectively compensate the magnetometer’s measurement values, and that the IMOA method more easily jumps out of the local optimum, and has higher compensation accuracy.