ELEC-E8107 - Stochastic models, estimation and control D, Lecture, 6.9.2022-7.12.2022
This course space end date is set to 07.12.2022 Search Courses: ELEC-E8107
Lecture 7, Nov 1: USE OF EKF FOR SIMULTANEOUS STATE AND PARAMETER ESTIMATION, MULTIPLE MODELS
USE OF EKF FOR SIMULTANEOUS STATE AND PARAMETER ESTIMATION
The EKF can be used to estimate simultaneously
• the base state and
• the unknown parameters of a system.
This is accomplished by stacking them into an augmented state and
carrying out the series expansions of the EKF for this augmented state.
MULTIPLE MODELS
The multiple model (MM) algorithms; Hybrid systems — the system behaves according to one of a finite number of models, it is in one of several modes (operating regimes), both discrete (structure/parameters) and continuous uncertainties
1. The static MM algorithm —for fixed (nonswitching) models
2. The optimal dynamic MM algorithm — for switching models —Markov chain, two suboptimal approaches: Generalized pseudo-Bayesian (GPB) ; Interacting multiple model (IMM)