An automatic real-time mode-matching MEMS gyroscope with fuzzy and neural network control

This paper reports a novel method to accomplish automatic and real-time mode-matching control for a MEMS vibratory gyroscope based on fuzzy and neural network algorithms. Experimental results demonstrate that it only needs about 8 seconds to fulfil mode-matching automatically in the fuzzy control system, and a mismatching error lower than 0.32Hz is achieved over the temperature range from -40°C to 80°C in the neural network real-time control system. The scale factor of the mode-matched gyroscope with the closed loop controlled sense mode is measured to be 65.9 mV/deg/s with nonlinearity about 0.03%, and the bias instability and the angle random walk (ARW) are evaluated to be 0.68 deg/h and 0.028 deg/h^1/2, respectively.