Unconstrained Optimization using the Extended Kalman Filter

A function using the extended Kalman filter to perform unconstrained nonlinear optimization

Description

The Kalman filter is actually a feedback approach to minimize the estimation error in terms of sum of square. This approach can be applied to general nonlinear optimization. This function shows a way using the extended Kalman filter to solve some unconstrained nonlinear optimization problems. Two examples are included: a general optimization problem and a problem to solve a set of nonlinear equations represented by a neural network model.

This function needs the extended Kalman filter function, which can be download from the following link:
http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=18189&objectType=FILE

Acknowledgements

The author wishes to acknowledge the following in the creation of this submission:
Learning the Extended Kalman Filter
This submission has inspired the following:
Nonlinear least square optimization through parameter estimation using the Unscented Kalman Filter

• MATLAB release: MATLAB 7.5 (R2007b)

[sudheer ch]

Most of the times gets caught with local minima. it needs a lot of improvement.

[V. Poor]

 

[Rohit Hippalgaonkar]

Hi I am looking for an example where the EKF is applied to a continuous-time non-linear system with non-zero inputs (say measurements are taken at regular time samples through a non-linear (even linear would do) measurement process. I have looked around for this kind of example in the standard texts but haven't found any.

Also a good source showing the implementation of the EKF wherein we linearize about a single operating point (as against linearizing about the predicted state every time) would be really helpful!

Thanks in advance!
Rohit