This paper proposes a new sigma point selection strategy to better capture the information of a probability distribution. By doing so, the non-local sampling problem inherent in the original unscented transformation (UT) is fundamentally eliminated. It is argued that the improved UT (IUT) outperforms the original UT at the cost of increased but comparable computation burden and will be useful in constructing a nonlinear filter. © Springer-Verlag Berlin Heidelberg 2004.
CITATION STYLE
Wu, Y., Wu, M., Hu, D., & Hu, X. (2004). An improvement to unscented transformation. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3339, pp. 1024–1029). Springer Verlag. https://doi.org/10.1007/978-3-540-30549-1_96
Mendeley helps you to discover research relevant for your work.