Research on Multi-Inertial Navigation Information Fusion Method Based on Kalman Filter

The error propagation characteristics of two or more inertial navigation system (INS) devices are different with different accuracy levels. When the high precision inertial navigation system fails, how to make the low precision inertial navigation system work independently in the best state is an important research content of multi-inertial navigation systems. The Kalman filter is used to fuse the information of strapdown inertial navigation system and rotating inertial navigation system. The common fusion methods are position fusion, velocity fusion, “velocity + attitude” fusion, and “position + velocity” fusion. Position and velocity are the linear motion information of ship navigation, which are slow variables, and the amount of error caused by transmission delay is small. Attitude is the angular motion information of ship motion, which is a fast variable. The error caused by transmission delay is large, and the tracking effect of azimuth error is poor under the condition of ship maneuvering. In this paper, the principle of multi-inertial navigation system information fusion based on Kalman filter is studied, and the state equations and observation equations of position fusion and position + velocity fusion of multi-inertial navigation system are further established. The correctness of the fusion scheme is verified by the fusion simulation of strap down and single-axis rotating inertial navigation system, and strap down and biaxis rotating inertial navigation system.