Estimating the Location and Orientation of Buried Objects Using the Magnetic Field Information of a Dipolar Response

Abstract

Two detection approaches of unexploded ordnance (UXO) are examined, both using magnetic field data collected in the electromagnetic induction regime and using the dipole approximation as a forward model. The first approach attempts to invert for position and moment from the knowledge of the magnetic field only, from which both the vector and scalar potentials are estimated. The method, analytic in nature, avoids the necessity of solving the ill-conditioned problem typically associated with UXO detection. Although successful, the generalization of this method to multi-UXO is not trivial. To address the multi-target configuration, we propose an alternative method based on the Gauss-Newton iterative algorithm. The latter, more numerically involved, can be formulated with -- in principle -- an arbitrary number of targets. Initial validating results are shown here with two targets which are successfully located and identified from synthetic data with varying degrees of noise.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 26, 2012
Accession Number
ADA577286

Entities

People

  • Tomasz M. Grzegorczyk

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Buried Objects
  • Contracts
  • Department Of Defense
  • Detection
  • Detectors
  • Dipole Moments
  • Electromagnetic Induction
  • Engineering
  • Equations
  • Grids
  • Magnetic Fields
  • Mathematics
  • Military Research
  • Standards
  • Unexploded Ammunition

Readers

  • Computational Modeling and Simulation
  • Linear Algebra
  • Plasma Physics / Magnetohydrodynamics