Semester

Fall

Date of Graduation

2012

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

James Smith

Committee Co-Chair

MaryAnn Clarke

Committee Member

Powsiri Klinkhachorn

Committee Member

Roy Nutter

Committee Member

Franz Pertl.

Abstract

Near-field magnetic principles and properties have been well studied and are used in a plethora of modern applications, ranging from medical applications to audio and video processing, and magnetic tracking. Current tracking applications are based in either AC or Pulsed-DC systems. Generally, AC systems have high resolution and accuracy, but perform very poorly in the presence of conducting magnetic materials. Pulsed-DC tracking has the benefit of not inducing large eddy currents in proximity to magnetic materials, thus increasing its overall accuracy. It has been suggested that pure DC systems are not feasible because they are unable to account for the presence of the Earth's magnetic field.;It was the purpose of this research to propose and create a system, and develop an algorithm, that has the ability to determine the three-dimensional position and orientation of a permanent magnetic source; the position and orientation to be determined by information reported by a network of single-axis magnetic sensors. Methodology to account for the Earth's magnetic field before, during, and after operation in order to remove ambient and environmental magnetic noise, much like a pulsed-DC system does, was also to be considered.;A center-finding algorithm was developed to determine position (x- and y-axis) based on the unique geometry of the B-field of the magnetic source at any point in three-dimensional space. Two degrees of orientation, elevation and rotation, were calculated from the position and the reported values of the magnetic sensors. The z-axis position was then determined given the analytical model and the other calculated values. In addition to the computed position, a six input Kalman tracker-estimator was developed and implemented using three dimensions of position and velocities to aid in predicting the path the magnetic source will take, based solely on kinematics, to reduce position-based sensor error.;The contribution of this research shows that is it not necessary to obtain three-axis magnetic data to track a magnetic source in three-dimensional space. When the distribution of the magnetic flux density is known, it is possible to determine three-dimensional position and orientation with only single-axis information.;Experimental testing verified the theoretical predictions of this statement. A rotational test apparatus was used to verify two-dimensional position and orientation, while a linear test apparatus verified position in three dimensions. The same magnetic source was used, while changing the orientation for each test. Initial findings allow the magnetic source to be tracked on the rotational testing apparatus to within a radial error of 3.9% (mean) and less than 6.4% (worst case) for predictions. The linear apparatus is able to track the z-axis component of the source which can be determined within 0.19% (mean) and 0.24% (worst case), and mean three-dimensional position of the magnetic source within 1.4% error. These results suggest that the novel method presented in this document is credible method for magnetic detection and tracking.

Download

Share

COinS