An Embedded Controller with Coded Signal Processing for Manipulating a Multi-Degree Piezostack-Driven Nanostage

摘要

The object of this research is to develop an embedded control strategy for controlling a prototyped real-time multi-degree piezoactuator-driven nanostage system, which consists of piezostack actuators, strain gauge sensors, nanostage mechanisms, high-voltage driving as well as high-precision sensing circuits, and high-resolution positioning control strategies based upon adaptive fuzzy control principles. Since the piezoacruators employed for driving the nanostage contain high nonlinearities, thousands of input/output data experienced by the nanostage system are acquired and identified in this research to establish a system estimator, which can mimic the real system input/output response behaviors for any input signals with a frequency ranging from 4 Hz to 10 Hz. The developed estimator model is then utilized to serve controller exercising. Since feedback signals of the flexible mechanisms are minute and sensitive to environmental interferences and/or background noises, this research proposes a novel approach to filter out high frequency noises lodged in the signals by implementing a moving average wavelet transform coding, in addition to the traditional shielding practices employed on the feedback paths.