Aiming at the problem that the turning rate of ship measured by MEMS gyroscope under the laboratory conditions fluctuated greatly

preprocess the original signal by average filtering and correct zero offset

and designed Kalman filter is used to realize further filtering

experiments show that once turning rate of the ship changed

the Kalman filter is no longer applicable due to response lag

then improve the filter and propose a simple and effective adaptive Kalman filtering method. Experimental results prove that the designed adaptive Kalman filter is effective and feasible

it can achieve data filtering while satisfying the dynamic performance.