courseraИнженерия

Linear Kalman Filter Deep Dive (and Target Tracking)

Курс от University of Colorado System

Средний≈ 21.7 чАнглийский

О курсе

As a follow-on course to "Kalman Filter Boot Camp", this course derives the steps of the linear Kalman filter to give understanding regarding how to adjust the method to applications that violate the standard assumptions. Applies this understanding to enhancing the robustness of the filter and to extend to applications including prediction and smoothing. Shows how to implement a target-tracking application in Octave code using an interacting multiple-model Kalman filter.

Навыки, которые вы освоите

Estimation Applied Mathematics Mathematical Modeling Linear Algebra Forecasting Statistical Methods Matlab Statistical Inference Correlation Analysis Time Series Analysis and Forecasting Numerical Analysis Markov Model

Программа курса

4 модулей · 94 учебных материалов

01Deriving the linear Kalman filter26 материалов

2.1: Introduction to the course

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

Учитесь у экспертов

Gregory Plett

Professor

Linear Kalman Filter Deep Dive (and Target Tracking)

О курсе

Навыки, которые вы освоите

Программа курса

2.1: Introduction to the course

Учитесь у экспертов

Gregory Plett

2.1.1: Welcome to the course!

2.1.2: Predict/correct mechanism of sequential probabilistic inference

2.1.3: The Kalman-filter gain factor

2.1.4: Summarizing the six steps of generic sequential probabilistic inference

2.1.5: Deriving the three linear Kalman-filter prediction steps

2.1.6: Deriving the three linear Kalman-filter correction steps

2.1.7: Summary of "Deriving the linear Kalman filter" module plus next steps

2.2.1: How do we improve the numeric robustness of a Kalman filter?

2.2.2: How do we increase the precision of the linear Kalman filter?

2.2.3: How do I initialize and tune a Kalman filter?

2.2.4: What do we do when the noises are nonzero-mean?

2.2.5: What do I do if the process and sensor noises are cross-correlated?

2.2.6: What about when the process and sensor noises are not white?

2.2.7: Summary of "Making the linear Kalman filter bulletproof" module plus next steps

2.3.1: Automatically detecting bad measurements with a Kalman filter

2.3.2: Processing measurements sequentially for multi-output systems

2.3.3: Using the Kalman filter for prediction

2.3.4: Using the Kalman filter for smoothing

2.3.5: Steady-state Kalman filters

2.3.6: Continuous-time Kalman filters

2.3.7: Summary of "Extensions and refinements to linear Kalman filters" module plus next steps

2.4.1: Some unique features of the target-tracking application

2.4.2: Tracking with polar measurements and a Cartesian state

2.4.3: The interacting-multiple-model Kalman filter

2.4.4: Implementing the IMM Kalman filter in Octave

2.4.5: Steady-state alpha-beta-gamma target-tracking filters

2.4.6: Summary of "Target-tracking application using a linear Kalman filter" module plus next steps