I’m taking this course in Q3 during my first year at Twente. The aim is to learn how to design optimal estimators that can uncover the time-dependent states of a dynamic system. It includes state-of-the-art architectures
- Kalman Filter for linear and Gaussian systems
- Extended Kalman Filter for near-linear and near-Gaussian systems
- Particle Filter for non-linear and non-Gaussian systems
The accent of the course is on the utilization aspects, not on a mathematically rigorous treatment of the topic. One of the challenges addressed in the course is how to bring theoretical concepts to a practical solution.
Topics covered so far
- Introduction to Optimal Estimation and Dynamics
- The Estimation Paradigm (the static case)
- Generalized Normal Distribution
- Fundamentals of parameter estimation - Part I, Fundamentals of parameter estimation - Part II, Fundamentals of parameter estimation - Part III
- Prediction in a linear dynamic system
- Discrete Kalman Filtering for Radar Tracking Applications
- Extended Kalman Filtering (EKF)
- Particle Filter, Particle Filtering
Sadly, we use MATLAB.
References:
- van der Heijden et al: Classification, Parameter Estimation and State Estimation, 2004, J Wiley.