System design

From OpenEMT.org
Jump to: navigation, search

Anser EMT is an open-source electromagnetic navigation platform for research use. The system can be readily build from the design files provided in this project. This page provides a brief summary of the system design. Detailed information can be found by clicking the individual section headings.

The Anser system is comprised of eight sub-systems summarized below:


Overview

System.PNG

Signal generator

This provides the source signal for the emitter coils of the field generator. Eight individual waveform generators ICs produce a set of sinusoidal signals in to 20kHz band for amplification.

AC current amplifier

A closed loop AC amplifier increases the power gain of the generated sinusoidal signals. This power gain allows the signal to drive the eight emitter coils to produce a time-varying magnetic field.

Tracking sensor

This coil produces a time-varying voltage signal when placed in the working volume of the field generator. The signal is a superposition of voltages induced by the time-varying magnetic field from each emitter coil.

Sensor amplifier

The induced sensor voltage is very weak, on the order of microvolts, and requires filtering and amplification before being sampled. This instrumentation amplifier circuit conditions the signal in preparation for sampling.

Data acquisition

Once conditioned, the tracking sensor is sampled by a data acquisition unit (DAQ). A current sense signal from the emitter coils is also sampled in order to determine the relative phase angle sensor voltage. Anser uses a National Instruments DAQ solution, but alternative/custom solutions may be used instead.

Signal demodulation

The tracked sensor signal and emitter current sense signals are both superpositions of eight frequency components, with each component due to a single emitter coil. Synchronous demodulation in software allows the magnitude and phase of each frequency component to be measured.

Position and orientation algorithm

The magnitude and phase information of each frequency components is fed into the position and orientation algorithm.