Research On Magnetic Target Localization Algorithm Based On Least Square Criterion And Kalman Filter

Magnetic object location and tracking technology based on magnetic anomaly play an important role in submarine detection,underwater monitoring,prospecting,robot location and tracking and other fields.In this technology,the magnetic target is usually regarded as a magnetic dipole.By solving the least square problem,the position coordinates,magnetic moment vector and other relevant parameters of the magnetic target are obtained to realize the positioning and tracking.However,the noise of magnetic sensors and magnetic interference of environment will affect the measurement accuracy of magnetic anomaly,and then affect the positioning accuracy.The main work of this project is to study the relevant models and algorithms to compensate and correct the positioning results according to the characteristics of noise and error in the actual measured magnetic anomaly vector.Magnetic anomaly location can be reduced to solving the least square problem.According to the theoretical calculation value of the magnetic dipole model and the measured value of the sensor array output,the objective function to be optimized is obtained according to the least square criterion.By solving the global optimal value of the objective function,the magnetic target can be located.Due to the limited measurement accuracy of the magnetic anomaly vector,there is still a big deviation between the global convergence result of the objective function and the actual situation,and it is difficult to reduce the deviation from the perspective of the solution algorithm.To solve this problem,based on the convergence of the objective function,this paper studies and designs a linear Kalman filter,which compensates and corrects the convergence results,and achieves a more stable positioning effect.Starting from the magnetic dipole model,this paper describes the establishment process of the objective function and the global optimal algorithm.Aiming at the influence of noise on the convergence result of objective function,a linear Kalman filter model is established to compensate and modify the location result.Numerical simulation and experimental results show that compared with the positioning results before correction,the fluctuation range of the final positioning results is smaller and more stable after compensation and correction by the linear Kalman Filter.