Auto Regressive Moving Average (ARMA) Modeling Method for Gyro Random Noise Using a Robust Kalman Filter.

Area of Science:

• Control Engineering
• Signal Processing
• Inertial Navigation Systems

Background:

• Conventional Autoregressive Moving Average (ARMA) modeling for gyro random noise suffers from slow convergence and requires substantial sample data.
• Accurate modeling of gyro noise is crucial for precision in inertial navigation and control systems.

Purpose of the Study:

• To develop a novel ARMA modeling method for gyro random noise that overcomes the limitations of conventional techniques.
• To enhance the speed and accuracy of ARMA parameter estimation using robust Kalman filtering.

Main Methods:

• The proposed method integrates ARMA model parameters as state variables within a robust Kalman filtering framework.
• It employs unknown time-varying estimators for observation noise to accurately determine its mean and variance.
• Robust Kalman filtering is utilized for precise estimation of ARMA model parameters.

Main Results:

• The developed ARMA modeling method demonstrates rapid convergence compared to traditional approaches.
• The technique achieves high accuracy in estimating ARMA model parameters for gyro random noise.
• A significant reduction in the required sample size for effective modeling was observed.

Conclusions:

• The robust Kalman filtering-based ARMA modeling method provides a faster and more accurate solution for gyro random noise.
• This advancement is particularly beneficial for applications demanding efficient and precise ARMA modeling with limited data.
• The method enhances the reliability and performance of systems relying on accurate gyro data processing.