Strapdown Inertial Measurement Method Based on Data Fusion Technique and Its Application to Railway Track

To further increase the accuracy of strapdown inertial measurement system, based on Kalman data fusion technique, a novel strapdown inertial measuring method for railway track geometrical parameters was proposed. By analyzing major error factors that affect the strapdown inertial system, the error state space equations were constructed. The selection of external observations and their influences on the observability and the degree of observability of the Kalman filter were also analyzed. The velocity and roll angle were used as external observations, and the random vibration interference was modeled with second-order auto-regression equations. The simulation results show that compared with the common Kalman filtering algorithm, the proposed method can greatly increase the estimation accuracy of Kalman filtering state variables, and the standard deviation of pitch angle and azimuth angle is decreased by 1.5° to 2.5°.