Non-Linear Dynamics Identification Using Gaussian Process Prior Models Within A Bayesian Context

Source Codes Library

Hamilton Institute, National University of Ireland, Maynooth & Industrial Control Centre, University of Strathclyde

Several codes are developed during the course of this research. These source codes are mainly available in either MATLAB or C (MEX-C) language. The codes are organised according to toolboxes; that are,

  1. Gaussian Process Toolbox 1.0
  2. Gaussian Process Toeplitz Toolbox 1.0
  3. Gaussian Process Schur Toolbox 1.0
  4. Gaussian Process SSTS Toolbox 1.0
  5. Multiple Gaussian Processes Toolbox 1.0

These are labelled as according to the development of the chapters in the thesis. These toolboxes are, respectively, contained in the following folders,

These source codes are also available on the Internet, from the following website, http://www.hamilton.ie/keith/research.htm. Interested users who are new to Gaussian process can take a look at the demonstration that is made available in GP Toolbox.

If you are using the source codes for personal use, do make sure to read the Disclaimer, Limitation and Liability below.

 

Requirement

These M-codes are developed using MATLAB version 7.0, with the C source codes compiled using GCC (in Linux) and Microsoft Visual C++ 6.0 (in Windows). Note that the codes have not been tested with version earlier than MATLAB 7.0.

 

Instruction to Compile Mex-C Codes

Mex-C source codes can be treated the same way as C codes. In order to compile these Mex-C source codes (.c files) to your preference, an appropriate C compiler is required. To simply compile a source code, e.g., xyz.c, just call the following command from the MATLAB environment.

>> mex xyz.c

 

Disclaimer

The source codes are provided "AS IS" without warranty of any kind. The functions that are contained in the source codes are not warranted to meet users' requirement or that the operationg will be uninterrupted or error-free.

 

Limitation and Liability

The author shall not be responsible for infringements of third parties rights. In no events, unless required by applicable law, shall the author be liable for any direct, indirect incidental, special, exemplary or consequential damages of any character including, without limitation, damages for loss of good will, work stoppage, computer failure or malfunction, or any and all other damages or losses, even if advised of the possibility of such damage. Also, the author is under no obligation to maintain, correct, update, change, modify or otherwise support these source codes.