Model-on-Demand Toolbox

The Model-on-Demand Model Predictive Control Toolbox was developed by Martin W. Braun, Anders Stenman, and Daniel E. Rivera, based on the dissertation research of Anders Stenman (Ph.D., Linköping University, 1999) and Martin W. Braun (Ph.D., Arizona State University, 2001).  The MoDMPC toolbox allows the user to efficiently evaluate data properties for nonlinear modeling, validate user decisions for Model-on-Demand estimation, and incorporate the estimator in a GPC type predictive controller to assess control performance. The package is MATLAB-based and consists of:

  • An interactive GUI for visualization and validation of MoD models
  • A Simulink library of MoDMPC controllers
  • A context sensitive web-based help file
  • A step-by-step web-based demo file

For access to the MoDMPC toolbox, please download the zip file from this link.  This version of the toolbox has been tested in Releases 2007b, 2008b, and 2009b of Matlab/SIMULINK. The code is provided completely free of charge on an "as is" basis.  By requesting the package, you explicitly agree to our disclaimer.

Papers describing the use of MoD and MoDMPC in our laboratory can be found here.