Initial Self-Alignment for Marine Rotary SINS Using Novel Adaptive Kalman Filter

Fujun Pei,Li Zhu,Jian Zhao

doi:10.1155/2015/320536

Abstract

The accurate initial attitude is essential to affect the navigation result of Rotary Strapdown Inertial Navigation System (SINS), which is usually calculated by initial alignment. But marine mooring Rotary SINS has to withstand dynamic disturbance, such as the interference angular velocities and accelerations caused by surge and sway. In order to overcome the limit of dynamic disturbance under the marine mooring condition, an alignment method using novel adaptive Kalman filter for marine mooring Rotary SINS is developed in this paper. This alignment method using the gravity in the inertial frame as a reference is discussed to deal with the lineal and angular disturbances. Secondly, the system error model for fine alignment in the inertial frame as a reference is established. Thirdly, PWCS and SVD are used to analyze the observability of the system error model for fine alignment. Finally, a novel adaptive Kalman filter with measurement residual to estimate measurement noise variance is designed. The simulation results demonstrate that the proposed method can achieve better accuracy and stability for marine Rotary SINS.

Highlights

Because Strapdown Inertial Navigation System (SINS) has special advantages, it has been widely used in aviation, marine, and land vehicle navigation and positioning
Because the initial errors in the derived navigation parameters cannot be depressed by rotary modulation technique, initial alignment is the key technique for Rotary SINS
The initial attitude angle can be obtained by initial alignment, which is divided into two procedures: coarse and fine alignment

Summary

Introduction

Because SINS has special advantages, it has been widely used in aviation, marine, and land vehicle navigation and positioning. In [11], the fine alignment error model in the inertial frame was developed, the velocity differences between the calculated Strapdown INS values and the true values along the inertial frame were considered as measurements, and Kalman filter was used to estimate the error angle. In this method, the disturbed acceleration was ignored based on the integration of the disturbed acceleration which was considered equal to zero. The s frame varied with the changes of IMU’s position is a real-time variable frame

Realization of Coarse Alignment for Rotary SINS in the Inertial Frame

Fine Alignment Error Model and Observability Analysis

Novel Adaptive Kalman Filter Designed for Fine Alignment

Simulation Results

Conclusions