Abstract

Many single-point spacecraft attitude determination algorithms are developed as solutions to Wahba’s problem, a weighted least squares problem for attitude determination from vector observations. When the weights assigned to the vector observations vary widely, for example, when coarse attitude sensors are coupled with a very fine attitude sensor, some solutions will encounter numerical difficulties. When the vector observations in both the reference and body frames are corrupted by noise with sophisticated models, which is the case with certain attitude determination problems arising from relative navigation, solving Wahba’s problem will not yield the statistically optimal attitude estimate. This talk presents two alternatives to Wahba’s problem that address the two issues, respectively. Also presented are algorithms that solve the two problems, their covariance analysis and relationships with existing algorithms, and numerical tests. The first problem is intended for scenarios with one dominant vector observation, where the other observations can be vector observations, arc-length observations (from GPS carrier phase differences, for example), and/or other data types. This problem has an efficient solution involving no iterative calculations. The second problem is formulated as a nonlinear total least squares problem and suited for the most general vector noise models.

Bio
Yang Cheng is an Associate Professor in the Department of Aerospace Engineering at Mississippi State University. His research interests include spacecraft GNC, space situational awareness, and nonlinear filtering. He received his B.Eng., M.Eng., and Ph.D. degrees in 1997, 1999, and 2003, respectively, all from Harbin Institute of Technology (China). He is a member of the AIAA GNC Technical Committee (2010–present) and an Associate Fellow of AIAA (2012–present). He has been an Associate Editor of The Journal of the Astronautical Sciences since 2011 and served as the Technical Discipline Chair of the 2018 AIAA GNC Conference.