I know that kalman uses the LMS criterion in its optimization step to reduce error.
But under certain conditions (e.g., deterministic inputs), the value of the estimation could be the same for Kalman and LMS as an algorithm (not only as a criterion used in Kalman). Can you explain for me why and how ? Preferred in words instead of equations
Thank you
Thank you Mr. Jagan for your explanation.
Thank you Prof. Zayyani for your paper. Actually it was my reference in my readings, and what I wrote in the questions was derived from this paper, but wanted a brief intuitive explanation in some words, on how are they related not only in the deterministic identification setting, but in a general way i.e., including also the stochastic case.
Georges, the Kalman filter may be considered as a generalization of the least squares technique to dynamical systems.
The classical least squares estimator exists in two equivalent forms, "batch" and "sequential". The equations of the sequential least squares estimator are the same as of the Kalman filter, except that the system dynamics matrix is identity and the process noise covariance matrix is zero. Thus, no prior information regarding the system dynamics is used for estimation. In the Kalman filter, this information is accounted for.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude