Large-Scale Magnetic Target State Estimation Using Model Parameter Optimization

Abstract Improved grey wolf optimizer-conjugate gradient least squares (IGWO-CGLS) algorithm is proposed to solve the state estimation problems of large-scale magnetic target location, target recognition and magnetic moment estimation. To address the state estimation problem of large-scale magnetic targets, a magnetic dipole array magnetic field observation model is established based on a single magnetic vector sensor, and the magnetic target model parameter inversion based on IGWO-CGLS is used to obtain the state of the magnetic target. The state of the target is obtained by inverting the parameters of the magnetic target model. Firstly, the signal preprocessing is carried out by thresholding and down sampling to extract the complete magnetic field passing characteristics, and then the IGWO is used to optimize the magnetic target position, speed, heading and length, and then the magnetic moment parameter is calculated by CGLS to obtain the state parameters of the magnetic target. And numerical experiments are designed to sublicense the influencing factors of state estimation accuracy. The numerical simulation results show that IGWO-CGLS has higher state estimation accuracy than particle swarm optimization-CGLS, artificial hummingbird algorithm-CGLS, artificial rabbits optimization-CGLS, GWO-CGLS, IGWO-stepwise regression. The ship model test results show that IGWO-CGLS can estimate the state parameters of the ship model better.