Integration of single-frame and filtering methods for nanosatellite attitude estimation

Single-frame and filtering algorithms are two standalone techniques for attitude estimation. Single-frame methods use the measurements obtained at a single point in time and fuse the information from sensors to optimally determine the attitude of the spacecraft. On the other hand, the filtering methods use the spacecraft dynamics information together with the measurements obtained over a period of time to sequentially estimate the spacecraft's attitude. Filtering methods are usually capable of providing more accurate attitude estimates than the single-frame methods and can provide estimates even when there are insufficient measurements for single-frame methods to work. In this chapter, we investigate different nanosatellite attitude estimation algorithms for which the single-frame and filtering methods are integrated and used together. First the single-frame method minimizes the Wahba's loss function to find the optimal solution for the attitude on the basis of magnetometer and sun sensor vector measurements. Then the filtering algorithm uses the attitude estimates of the single-frame method as measurements and provides more accurate attitude information as well as estimates for the additional states.