Search code examples
covariancegaussiankalman-filter

How can I compute the sigma points for UKF?


https://towardsdatascience.com/the-unscented-kalman-filter-anything-ekf-can-do-i-can-do-it-better-ce7c773cf88d

Image above & tutorial: https://towardsdatascience.com/the-unscented-kalman-filter-anything-ekf-can-do-i-can-do-it-better-ce7c773cf88d

I am confused about how to compute the sigma points for the unscented Kalman filter. For me, mu is a 2-dim vector, so n is 5 and cov is a 2x2 matrix. lambda is 3-n, so 1. Now, I don't understand the index i, since non of the values/matrices are dependent on i. What would be the difference between X[1] and X[2]?

Thanks for any help, I think I'm probably just confused.. :)


Solution

  • I realized i simply stands for the i'th columnn.