Home » Analysis and Comparison of Extended and Unscented Kalman Filtering Methods for Spacecraft Attitude Determination by Orlando X. Diaz
Analysis and Comparison of Extended and Unscented Kalman Filtering Methods for Spacecraft Attitude Determination Orlando X. Diaz

Analysis and Comparison of Extended and Unscented Kalman Filtering Methods for Spacecraft Attitude Determination

Orlando X. Diaz

Published
ISBN :
Kindle Edition
97 pages
Enter the sum

 About the Book 

Two methods of estimating the attitude position of a spacecraft are examined in this thesis: the extended Kalman filter (EKF) and the unscented Kalman filter (UKF). In particular, the UnScented QUaternion Estimator (USQUE) derived from [4] isMoreTwo methods of estimating the attitude position of a spacecraft are examined in this thesis: the extended Kalman filter (EKF) and the unscented Kalman filter (UKF). In particular, the UnScented QUaternion Estimator (USQUE) derived from [4] is implemented into a spacecraft model. For generalizations about the each of the filters, a simple problem is initially solved. These solutions display typical characteristics of each filter type. The UKF is very attractive in spacecraft attitude estimation, given that spacecraft dynamics are highly nonlinear. For nonlinear systems, the UKF is of particular interest because it uses a carefully selected set of sample points that more accurately map the probability distribution than the linearization of the standard extended Kalman filter. This leads to faster convergence of the attitude solution from largely inaccurate initial conditions. The filter created in this thesis is formulated based on Markley and Crassidis’s work on standard attitude-vector measurements using a gyro-based model for attitude propagation. From the standard attitude vector measurements, the global attitude parameterization is found and given by a quaternion, while a generalized three-dimensional attitude representation is used to define the local attitude error. The multiplicative quaternion-error is then found from the local error. The simulation results indicate that the unscented filter is more robust than the extended Kalman filter.