Loading...

Adaptive Estimation of Satellite Attitude via an Entropy Approach

Barzegar, Aylin | 2017

645 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49716 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Pourtakdoust , Hossein; Kiani, Maryam
  7. Abstract:
  8. Nonlinear state estimation algorithms developed based on Bayesian framework have been increasingly utilized in space navigation and attitude estimation to enhance the mission performance. However, successful implementation of these algorithms requires accurate knowledge of process and measurement noise statistical properties. As the satellites experience various systematic and environmental uncertainties on orbit, no priori information of noise characteristics is usually available. In this regard, adaptive estimation algorithms have been widely utilized to tune the planned noise properties according to the variable working conditions. Among various adaptation methods, innovation-based algorithms are theoretically more popular, but they suffer from two operational limitations of inability to hold the positive definiteness property of noise matrices and strong dependency of filter performance on window size selection. In this regard, the current research is dedicated to study and propose new operational ways to deal with these difficulties. As such, noise covariance matrices are remained positive definite utilizing Cholesky factor decomposition method, and the window size has been adaptively adjusted based on the relative entropy and confidence level approaches. The proposed adaptation schemes have been applied to nonlinear filters of extended Kalman filter (EKF) and cubature Kalman filter (CKF) to provide an online adaptive estimation framework for attitude estimation of a low Earth orbit (LEO) satellite equipped with sun sensor and three-axis magnetometer measurements. The viability and performance of the proposed adaptive CKF (ACKF) and adaptive EKF (AEKF) are verified and demonstrated via Monte Carlo analysis. In addition, a sensitivity analysis on orbital elements, environmental and dynamic system characteristics is performed to verify the feasibility and utility of the proposed filters over a wider range of operating environment
  9. Keywords:
  10. Attitude Estimation ; Innovation ; Satellite Attitude Control ; Window Size Selection ; Relative Entropy ; Confidence Level

 Digital Object List

 Bookmark

No TOC