courseraИнженерия

Nonlinear Kalman Filters (and Parameter Estimation)

Курс от University of Colorado System

Средний≈ 21.1 чАнглийский

О курсе

As a follow-on course to "Linear Kalman Filter Deep Dive", this course derives the steps of the extended Kalman filter and the sigma-point Kalman filter for estimating the state of nonlinear dynamic systems. You will learn how to implement these filters in Octave code and compare their results. You will be introduced to adaptive methods to tune Kalman-filter noise-uncertainty covariances online. You will learn how to estimate the parameters of a state-space model using nonlinear Kalman filters.

Навыки, которые вы освоите

Estimation Matlab System Implementation Calculus Derivatives Control Systems Numerical Analysis Mathematical Modeling Statistical Methods Applied Mathematics Mathematical Software Model Evaluation Simulation and Simulation Software

Программа курса

4 модулей · 97 учебных материалов

01The extended Kalman filter30 материалов

3.1: Introduction to the course

Frequently Asked QuestionsЧтение Course ResourcesЧтение How to Use Discussion ForumsЧтение Introduce yourselfОбсуждение

Учитесь у экспертов

Gregory Plett

Professor

Nonlinear Kalman Filters (and Parameter Estimation)

О курсе

Навыки, которые вы освоите

Программа курса

3.1: Introduction to the course

Учитесь у экспертов

Gregory Plett

3.1.1: Welcome to the course!

3.1.2: Introducing nonlinear variations to Kalman filters

3.1.3: Deriving the three extended-Kalman-filter prediction steps

3.1.4: Deriving the three extended-Kalman-filter correction steps

3.1.5: Introducing a nontrivial EKF example, finding derivatives

3.1.6: Introducing Octave code to initialize and control EKF for state estimation

3.1.7: Introducing Octave code to update EKF for state estimation

3.1.8: Summary of "The extended Kalman filter" module plus next steps

Lesson 3.2.1: Problems with EKF that are improved with sigma-point methods

3.2.2: Approximating uncertain variables using sigma points

3.2.3: Deriving the six sigma-point-Kalman-filter steps

3.2.4: Introducing Octave code to initialize and control SPKF for state estimation

3.2.5: Introducing Octave code to update SPKF for state estimation

3.2.6: Summary of "The sigma-point (unscented) Kalman filter" module plus next steps

3.3.1: Iterating the EKF for systems having significant measurement-equation nonlinearities

3.3.2: Implementing the IEKF in Octave, plus an example

3.3.3: Simplifying the SPKF when noises are additive

3.3.4: Implementing the cubature Kalman filter (CKF) in Octave

3.3.5: Adaptively estimating noise covariance matrices: The AEKF

3.3.6: Implementing the AEKF in Octave

3.3.7: Summary of "Extensions and refinements to nonlinear Kalman filters" module plus next steps

3.4.1: Deriving SPKF method for parameter estimation

3.4.2: Implementing SPKF parameter estimation in Octave

3.4.3: Deriving EKF method for parameter estimation

3.4.4: Implementing EKF parameter estimation in Octave

3.4.5: How to estimate states and parameters at the same time

3.4.6: Implementing EKF joint state/parameter estimation in Octave

3.4.7: Summary of "Parameter estimation and joint estimation" plus next steps