DASL 108: Kalman Filtering

Author: Blake Hament Email: blakehament@gmail.com
Date: Last modified on 10/30/2018

This course is an introduction to Kalman Filtering. It begins with review of important background topics, then develops and explores the algorithms for linear and nonlinear Kalman Filtering. The last part of the course requires students to collect data with a noisy IMU and write their own Kalman Filters to estimate the ground truth.

The class is organized roughly as follows:

1. Background (4 hours): Statistics, Optimization, Linear Algebra, Bayesian Probability
2. Kalman Filtering and Extended Kalman Filtering (4 hours)
3. Practical Application (4 hours)

• Basic Dynamics [ME 230]
• Some Statistics and Linear Algebra
• Matlab or other coding language

Topic Outline: Kalman Short Course Outline

Excellent Beginner tutorial: https://home.wlu.edu/~levys/kalman_tutorial/

Probabilistic Robotics: http://www.probabilistic-robotics.org/

Types of Noise

Least Squares Regression

Variance and STD

Riccatti Equation

Bayesian Probability and Filtering

Kalman Filtering

Pre-Assessment

Exam