Closed-loop neural network controlled accelerometer

The purpose of this paper is to present aspects of an integrated micromachi ned sensor-neural network transducer development. Micromachined sensors exh ibit particular problems such as non-linear characteristics, manufacturing tolerances and the need for complex electronic circuitry. The novel transdu cer design described here, based on a mathematical model of the micromachin ed sensor, is aimed at improving in-service performance and facilitating de sign and manufacture over conventional transducers. The proposed closed-loo p transducer structure incorporates two modular artificial neural networks: a compensating neural network, which performs a static mapping, and a feed back neural network, which both linearizes and demodulates the feedback sig nal. Simulation results to date show an excellent linearity, wide dynamic r ange and robustness to shocks for the proposed system. The design was appro ached from a control engineering perspective due to the closed-loop structu re of the transducer.