Hybrid Intelligent Control: Implementation and Application

This paper describes how cognitive methods, such as Fuzzy Controllers, Arti cial Neural Networks, and Genetic Algorithms can be integrated together to produce a family of Hybrid Intelligent Controllers, which provide good performance in a variety of complex, real-time and non-linear problems. Two new paradigms are introduced, namely Weighted Radial Basis Functions and Hierarchical Hybrid Fuzzy Controllers, to exploit all the advantages of the proposed approach. A dedicated hardware, consisting of an ad-hoc VLSI chip, and its programming environment have been developed as a test bed for the proposed technique. A practical application is also presented.