Advanced Kalman Filter Theory and Practice

Update as of 3/14/04: Course Materials for a class given recently presented on "Kalman Filter Theory and Navigation Applications."
CyberStrategies, Inc. is pleased to offer two engineering short courses -- one elementary and one advanced -- on Kalman filter theory and applications. Both courses are five-days long and both use Matlab scripts and Simulink models to illustrate the key concepts. The elementary course presents the subject from a more practical, engineering perspective with a minimum of mathematical theorems and proofs. The advanced course offers a deeper understanding of the mathematical foundations of Kalman filter theory from a Hilbert Space and stochastic calculus perspective.
Regular Course Fee: $1995.00.
If you register at least 90 days before the scheduled date of the course, the cost is only $1595.00
Sign up at least 30 days before the scheduled date, and the cost is only $1795.00
Click here to register and pay with your credit card.
Otherwise, you can register and pay by check.
Or, if you just have questions, click here.
If you are new to Kalman filtering, an excellent place to start is the web site maintained by Greg Welch and Gary Bishop at the University of North Carolina. Overview of Advanced Course
Course Details
About the Instructor

The goal of this course is to provide practitioners with both a solid grounding in the mathematical concepts underlying Kalman filter theory and hands-on practice in applying these concepts to real world engineering problems.

Many practical examples are presented in this course. Students gain hands-on experience by working on real-world problems drawn from a variety of applications. The course also provides a collection of Matlab scripts and Simulink models illustrating the concepts and techniques of real-world Kalman filtering.

This course in unique in providing the student with many opportunities to participate in intensive individual and team problem-solving and exercise sessions. While other courses on this subject tend to render students passive by subjecting them lecture after dry lecture, this course allocates plenty of time for students to roll up their sleeves and actually apply the tools they learn.

Some of the advanced concepts include those of measure theory, Hilbert space theory (a light introduction), mean-square calculus, stochastic process theory, dynamic systems, and system identification. They are presented only as required to present a mathematically sound derivation of Kalman filter theory.

After taking this course, engineers should be in a position to analyze existing Kalman filters as well as to design and construct new ones. Special emphasis will be given to using spectral system identification techniques to handle non-linearities and augment the state vector. As a side benefit, after taking this course you will be able to read and understand the advanced literature of current research in the field of Kalman filtering.

This course is especially recommended for engineers with some experience in Kalman filtering who need to deepen their understanding of the underlying mathematical foundations and learn advanced modeling techniques for modifying existing filters. However, no prior knowledge of Kalman filtering is required in order to benefit from this course. The minimum prerequusites are a knowledge of calculus and some acquaintence with matrix algebra.
This course takes the perspective that a random process is a curve in Hilbert space and that optimal estimation is orthogonal projection in Hilbert space. Thus all the mathematics prerequisite to achieving this perspective are presented, starting with set theory, measure theory, etc.

For more detailed information on this course, and for an opportunity to register for the course, follow the links below.

[ Home | Overview | Course Details | About the Instructor | Register | Inquire ]







Kalman filtering measure theory Carroll advanced course

covariance estimation optimal control

Concord, California

Hilbert space

inertial navigation GPS



Ricatti Equation


Kalman filtering course