A Matlab/Simulink Toolbox for Inversion of Local Linear Model Trees

Models in the form of characteristic diagrams or more specifically, in the form of engine operating maps are mostly used in the automobile industry. This yields a large amount of measurements and involves the use of advanced instrumentations. This paper shows a developed software environment, namely a toolbox for the program Mat-lab/Simulink developed by company Mathworks. The name of the toolbox is "Inversion of the Local Linear Model Trees" and it basically consists of a local inversion of the Local Linear Model Trees (LOLIMOT). The importance of the inversion in control problems is widely known. Neural networks are a very effective and popular tools mostly used for modeling. The inversion of a neural network produces real possibilities to involve the networks in the control problem schemes. The developed toolbox is explained with the help of diagrams and GUI structure from Matlab which tend to clarify the idea of the program and its structure. The presentation is organized as a short tutorial of the toolbox, so that a potential user can directly understand how to access it. Nevertheless, formal mathematical equations, concerning the neural networks and membership functions, need to be explained together with the LOLIMOT structure. To validate and to clarify the explained toolbox, an example from a system used in the automobile industry is briefly shown.