Dpto. de Informática y Automática.

Information processing is a crucial topic in fuzzy systems, which we have studied from different points of view. First of all, we have remarked the importance of the procedure of knowledge acquisition, since the performance of Fuzzy Controllers are based on it. Some of the usual techniques applied to get this knowledge-and also the problems that can arise employing them- have been exposed. Studying the Fuzzy Logic Controller (FLC) configuration, we have penetrated in the blocks in those the knowledge is manipulated, showing the influence of the different component characterizations on the control action. The management of the information, both input and output, allows different specifications whose influence is shown by simulation examples. Finally, the utility of this information processing is shown since we can vary the control action without needing to change the FLC structure, allowing to reach the control aims and improving its performance