Attitude Determination System for a Cubesat Experiencing Eclipse

Area of Science:

• Aerospace Engineering
• Control Systems
• Signal Processing

Background:

• Kalman filters rely on predicted error covariance (Pk+1) and measurement noise (R) matrices to model uncertainties.
• These matrices can become inaccurate in real-world scenarios like CubeSat missions due to measurement faults.
• Accurate attitude estimation is critical for CubeSat operations and mission success.

Purpose of the Study:

• To develop an adaptive extended Kalman filter for CubeSat attitude estimation.
• To dynamically estimate the predicted error covariance matrix (Pk+1) and measurement noise covariance matrix (R).
• To improve attitude estimation accuracy, especially under challenging orbital conditions.

Main Methods:

• Implementation of an adaptive extended Kalman filter (AEKF).
• Utilizing an expectation-maximization (EM) algorithm for dynamic matrix estimation.
• Conducting simulation experiments to validate the AEKF's performance.

Main Results:

• The proposed AEKF demonstrated superior attitude estimation accuracy compared to existing methods.
• Performance improvements were particularly notable during sunlit and shadowed orbital phases.
• The algorithm achieved better accuracy with identical filtering parameters and initial conditions.

Conclusions:

• The adaptive extended Kalman filter effectively addresses time-varying uncertainties in CubeSat attitude estimation.
• Dynamic estimation of Pk+1 and R using the EM approach enhances filter robustness.
• This method offers a significant advancement for reliable CubeSat attitude determination.