Application of Improved 5th-Cubature Kalman Filter in Initial Strapdown Inertial Navigation System Alignment for...

Area of Science:

• Navigation Systems
• Estimation Theory

Background:

• The third-degree Cubature Kalman Filter (CKF) exhibits insufficient accuracy for initial alignment under large misalignment angles.
• Accurate initial alignment is critical for the performance of navigation systems.

Purpose of the Study:

• To propose an improved fifth-degree CKF algorithm for enhanced initial alignment accuracy.
• To address the limitations of existing CKF methods in scenarios with significant misalignment.

Main Methods:

• Recursive calculation of the innovation covariance matrix using an innovative sequence with an exponent fading factor.
• Development of a new adaptive error covariance matrix scaling algorithm.
• Application of Singular Value Decomposition (SVD) to improve numerical stability.
• Termination of the scaling scheme based on the azimuth gradient to prevent overshoot.

Main Results:

• The improved fifth-degree CKF algorithm demonstrates superior alignment accuracy compared to the traditional algorithm.
• The algorithm performs effectively even under large misalignment angle conditions.
• Enhanced numerical stability and controlled convergence were achieved.

Conclusions:

• The proposed fifth-degree CKF algorithm offers a significant improvement for initial alignment, particularly in challenging large misalignment scenarios.
• The integration of adaptive scaling and SVD enhances the robustness and accuracy of the filtering process.
• This advancement contributes to more reliable navigation system performance.