Abstract
To tackle the problem of high-accuracy self-calibration for marine dual-axis Rotary Inertial Navigation System (RINS) on moving base, this paper proposes a dynamic self-calibration approach with anti-disturbance capabilities. The primary work is twofold. Firstly, a state space model utilizing the Kalman filter is established, grounded in the error model of Inertial Navigation System (INS) and the calibration model of Inertial Measurement Unit (IMU). Secondly, a disturbance velocity elimination method based on wavelet transform that effectively eliminates disturbance velocity during calibration processes is proposed. Additionally, to tackle the low accuracy associated with boundary data in wavelet transform, a boundary extension method base on Particle Swarm Optimization (PSO) is introduced, which significantly enhances the accuracy of boundary data in wavelet transform. Experimental results indicate that the calibration accuracy of proposed method on static base are basically the same with on moving base. Gyro calibration accuracy is improved by over 92% relative to traditional non-anti-disturbance methods, moreover, other key parameters accuracy exhibits varying degrees of enhancement. Furthermore, a long-term maritime navigation experiment was conducted to validate these results, revealing an improvement in positioning accuracy by 24%, which significantly improved navigation accuracy of RINS under marine conditions.
Funding Information
  • National Key Research and Development Program of China (2023YFC2907000)

This publication has 29 references indexed in Scilit: