Introduction to Estimation and Kalman Filtering
29th October 2009, Glasgow

This meeting has since been held. (Register your interest for the next course here.)

Some of the presentations may now be available for download on-line (ACTC members only). You can do so from our Download Centre or simply by clicking on the appropriate link in the agenda below.

Please Note: Presentation material downloaded from our web-site should not be incorporated (either in part or in entirety) into other work etc., nor distributed (either in part or in entirety) to third parties, without the express permission of the authors.

Employees of ACTC member companies, are entitled to two places free of charge. Employees of companies that are not ACTC members will be required to pay a nominal fee to help defray costs

Overview

This course is aimed at introducing the Estimation theory, Kalman Filter and its application to engineers. Kalman filter is an efficient recursive filter that is capable to estimate a state from a series of measurement of the other states of a linear dynamic system. In order for parameter/state estimation of a nonlinear system, extended Kalman filter is used for this application and this is covered in this course.

The Kalman Filter and Extended Kalman Filter theory and practical applications are presented. Significant hands-on examples are used to reinforce the lectures.

Agenda
08.45 REGISTRATION
09.00 Introduction to Probability, Stochastic Processes and Signals (Basic Theorems, Disturbances & Noise Representation in Linear System)
09.45 Hands-on Session: Implementation of Disturbance & Noise in State-Space Model
10.45 TEA/COFFEE
11.00 Introduction to Kalman Filter (Continuous and Discrete Time)
12.00 Discrete Time Kalman Filter (Derivation, Properties, Riccati Equation and Tuning)
12.45 LUNCH
13.30 Hands-on Session: Application of Observers & Building the Kalman Filter
14.30 Introduction to Time Varying and Nonlinear Systems
15.00 TEA/COFFEE
15.15 Parameter Estimation using Extended Kalman Filters (Condition Monitoring, Model Based Fault Detection Methods)
16.00 Hands-on Session: Kalman Filtering for Parameter Estimation
17.00 CLOSE