Intelligent Control

Recommended Prerequisites

Background courses in the area of control. The course of “Control”.

Teaching Methods

A combination of the following methods will be employed: tutorial guidance classes and/or seminary; project of research and synthesis about a subject and/or of realization of practical assignments, and/or of research works, concerning simulation and/or real implementation.

Learning Outcomes

Acquire skills required to understand and apply the fundamental concepts concerning fuzzy systems, neural networks, clustering, support vector machines, and genetic algorithms, as well as to analyse, understand, and design fuzzy control systems and control systems based on neural networks. Acquire skills required to understand and apply the fundamental concepts concerning computational learning and computational intelligence. It is a general course with application in all the domains of engineering and computer science where the processing of data should be required.

Acquiring competencies in analysis and synthesis, autonomous learning, practical application of theoretical knowledge, problem solving, critical reasoning, adaptivity to new situations.

Work Placement(s)

No

Syllabus

Fuzzy sets. Operations and relations between fuzzy sets. Fuzzy systems. Inference systems. Neural networks. Neural network architectures. Computational learning. Supervised learning, unsupervised learning, and reinforcement learning. Generalization. Learning in neural networks. Fuzzy modelling. Learning methods for the synthesis and automatic tuning of fuzzy systems. Non-linear control based on fuzzy and neural models: non-adaptive control and adaptive control. Knowledge-based fuzzy control. Fuzzy supervised learning. Predictive control. Internal model control. Inversion of fuzzy systems and its application to control. Integration of neural networks and fuzzy systems. Neuro-fuzzy control. Clustering; Fuzzy clustering; Principal component analysis. Classification; Pattern recognition; Support vector machines; Evolving and genetic algorithms. Applications.

Head Lecturer(s)

Rui Alexandre de Matos Araújo

Assessment Methods

Assessment
Evaluation consists of a final exam with a weight of 50% on the final mark, and fieldwork or laboratory work, and/or synthesis work, and/or research work, with a weight of 50% on the final mark: 100.0%

Bibliography

• Wang, L.-X. (1997) – A Course in Fuzzy Systems and Control, Prentice-Hall.

• Haykin, S. (1994) – Neural Networks: A Comprehensive Foundation, Macmillan.

• Araújo, R. (2007) – Controlo Inteligente [Apontamentos], Departamento de Engenharia Electrotécnica e de Computadores, Faculdade de Ciências e Tecnologia da Universidade de Coimbra.

• Araújo, R. (2007) – Controlo Inteligente: Exercícios e Soluções, Departamento de Engenharia Electrotécnica e de Computadores, Faculdade de Ciências e Tecnologia da Universidade de Coimbra.

• Lin, C.-T., Lee, C.S.G. (1997) – Neural Fuzzy Systems - A Neuro-Fuzzy Synergism to Intelligent Systems, Prentice-Hall.

• Babuska, R. (1998) – Fuzzy Modeling for Control, Kluwer Academic Publishers.

• Krose, B., Smagt, P.V.D. (1996) – An Introduction to Neural Networks, The University of Amesterdam.