SUMMARY

This dissertation has been motivated by two practical applications: The first responds to a need to develop a geomagnetic field-based sensing system to help visually impaired patients locate objects and guide way-finding. The second is a method for intelligent manufacturing applications, where an eddy current is electromagnetically induced in a conductive workpiece for real-time measurements of geometrical features. This thesis addresses a common problem in these applications, which involves reconstruction of a physical field from limited measurements for characterizing geometrical features. Inspired by the simplicity of electromagnetic source-based models, this thesis has formulated the forward and inverse electromagnetic problems and derived computationally efficient closed-form solutions to provide a basis for developing a multi-function electromagnetic sensing system and reconstructing a physical field of interest. The source-based models have been employed in the design, analysis and optimization of the sensing system for the above-mentioned applications, upon which a prototype sensing system has been developed to serve as a test-bed for physical field reconstruction. An experimental investigation has been carried out to validate the concept feasibility of field reconstruction, verify the source-based models and evaluate the sensor performance. Apart from the development of a multi-function electromagnetic sensing system for geometrical feature measurements, the findings of this thesis have offered a basis of a general field-reconstruction method that has a spectrum of engineering applications.