Maxplus

The Maxplus toolbox of Scilab This toolbox has been developed by Michael McGettrick, Guy Cohen, Stéphane Gaubert, and Jean-Pierre Quadrat. M. McGettrick wrote the initial version, as part of his work within the ALAPEDES project funded by the TMR program of the European Community. A second version which is essentially the distributed one was developed by S. Gaubert and J.-P. Quadrat with the help of G. Cohen. The toobox includes fast numerical methods like a combinatorial Howard algorithm by S. Gaubert. It is currently maintained by J.-P. Quadrat. The integration in Scilab-gtk and ScicosLab is due to Jean-Philippe Chancelier. The max-plus toolbox of Scilab implements various algorithms (sum, product, Kleene star, residuation, eigenvalues and eigenvectors) for max-plus matrices. It allows one in particular to handle large sparse matrices. It also has an experimental part solving systems of max-plus linear equations and inequations. It is written in C and Fortran. Some functionalities are illustrated in the following tutorial. Distributions of the Maxplus toolbox, working for Scilab up to the 3.x.x series, can be found in the contrib section of the SCILAB distribution. A distribution adapted to the scilab4.x.x versions can be found here: Maxplus2.3.3c The maxplus toolbox is installed by default in ScicosLab, a new branch, coming from a Gtk+ version of Scilab (ScilabGtk), developed by some of the researchers who originally developed Scilab, it runs out of the box. If you do not have Scilab installed yet, or are ready to change for ScicosLab, this is probably the simplest way to get the toolbox.