Tracking Magnetic Target Based on Internative Multi-Model Square Root Unscented Kalman Filter

Abstract

Scalar magnetic sensors are widely used in magnetic detection and tracking because of their high sensitivity, wide detection range, and small attitude influence. Magnetic anomaly tracking usually uses scalar magnetometers arrays to retrieve the magnetic target position because the incomplete measurement of the geomagnetic field can’t be converted into a filtering problem. However, the workload of calculation is time-consuming, the target is easy to lose. Focusing on filtering divergent and targets easy to lose, tracking magnetic target based on Internative Multi-model Square Root Unscented Kalman Filter is proposed. Firstly, the magnetic anomaly signal is modeled based on moving magnetic target. Subsequently, the initial positioning of the magnetic target is based on dual-magnetometer trajectory planning. Finally, the IMM-SRUKF filter method is used to track magnetic targets adaptively. Both positioning and tracking of magnetic target based on IMM-SRUKF have been verified by simulation and experiments, the result shows that the magnetic target based on IMM-SRUKF can track target stably for a long time and adaptive to the target’s changed state, which indicates that the target is hard to lose.