Recursive Bayesian Method for Tracking a Magnetic Target with a Gradiometer
Abstract
This report describes a numerical method that may be used to efficiently locate and track magnetic targets with a tensor gradiometer. A target containing ferromagnetic material can be adequately modeled at a distance by an equivalent magnetic dipole. This magnetic target can be observed by means of a magnetic gradiometer that measures a symmetric, traceless gradient tensor as a function of time. Of interest is the inverse problem of the determination of the magnetic parameters of the target, and its position and velocity relative to the sensor at each time step. The previous method of direct inversion of the non-linear equations of the magnetic gradient tensor provided multiple solutions, and the results can be highly sensitive to noise in data. In this study, the determination of target magnetic moment, position and velocity is formulated as an optimal stochastic estimation problem, which could be solved using a sequential Monte Carlo based approach known as the particle filter . In addition to the conventional particle filter, the proposed tracking and classification algorithm uses the unscented Kalman filter (UKF) to generate the prior distribution of the unknown parameters. The proposed method is then demonstrated by using it to locate and track an automobile over a period of time using real data collected with a magnetic gradiometer. Two cases are investigated: (1) the observation contains only gradiometer data when a double solution exists, and (2) magnetic field components are added to the previous case and a unique solution is obtained. The automobile was moving either on a straight or a curved track.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2006
- Accession Number
- ADA475335
Entities
People
- Marius Birsan
Organizations
- Defence Research and Development Canada