KALMAN FILTERING APPLIED TO MEASUREMENTS IN ARMAMENT
Aim of the Course:
Purpose of the course is to gain knowledge in Kalman Filtering Applied to Measurement. The lectures will cover linear and extended Kalman filter and parameter estimation, but the course is organized through working examples rather than through theoretical explanation. Ten numerical examples prepared in Matlab are run to learn how to apply the Kalman filter, and to illustrate the possibility of the method. Also possibility of the Kalman filter will be compared with least squares technique. It is assumed that attendants of course have good knowledge in mathematics, computer skills, Matlab and general knowledge in measurement. The attendees will obtain the copy of the Matlab files. As option practical work is possible on Instrumented Physical Pendulum (Portable laboratory - see course outline for Pracitical Exercises in Measurements).
Who should attend?
The course is designed for graduated students, engineers - researchers in the research institutions. It is advanced course. The subject is very specific and multidisciplinary, so for the successful performing the course the students should have good basic knowledge in mathematics, physics, probability theory, statistics, system theory, computer skill and weapon systems.
Duration:
Duration is two weeks (twelve working days); 50 lectures (one lecture duration 45 min), but other arrangement is possible.
Course Outline
1. Review of Digital Kalman Filtering Techniques. Method of Least Squares vs. Kalman Filter
|
2. Polynomial Kalman Filters - Testing of accelerometer
|
3. State Estimation and Parameter Identification of Instrumented Physical Pendulum by Linear Kalman Filter
|
4. State Estimation and Parameter Identification of Instrumented Physical Pendulum by Extended Kalman Filter
|
5. State Estimation and Parameter Identification of Instrumented Physical Pendulum by Nonlinear Least Squares Method and comparison with Kalman filtering Techniques
|
6. Drag Identification from Doppler Radar Measurement by Extended Kalman filter
|
7. Tracking of Cannon Launched Projectile - Linear Decoupled Polynomial Kalman filter
|
8. Tracking of Cannon Launched Projectile - Linear Coupled Polynomial Kalman filter
|
9. Tracking of Cannon Launched Projectile - Extended Kalman filter
|
Lecturer: Dr Miodrag Curcin