ACS323 Intelligent Systems

Module Description

The last two decades or so have seen a real upsurge in the development and application(s) of the so-called intelligent systems based technologies, or as L.A. Zadeh referred to them as ‘soft-computing’, in several domains relating to modelling, signal processing, control and optimisation. This upsurge is mainly due to their ability to describe complex environments in a simpler way without compromising on either accuracy or transparency. Real-world environments which can be non-linear, multi-variable, noisy and characterised by ‘uncertainties’ can be dealt with effectively using this ‘intelligent’ formalism that ‘mimics’, artificially, the workings of the human ‘brain’, i.e. learning, adaptive, generalising and flexible. Hence, the objective of the course is to introduce students to the concepts of ‘machine-learning’ or ‘machine intelligence’ and ‘approximate reasoning’ (universal approximation) with a particular emphasis on Fuzzy Systems & Artificial Neural Networks and the related synergy of the Neural-Fuzzy architecture.

Credits: 10 (Autumn semester)

Module Leader

Mahdi MafoufProfessor Mahdi Mahfouf
Room D07, Amy Johnson Building

If you have any questions about the module please talk to me during the lectures or the labs in the first instance. It is likely that other students will learn from any questions you ask as well, so don’t be afraid to ask questions.

Outside of lectures please contact me via email, or drop in to see me.

Learning Outcomes

By the end of the module students will be able to:

  1. Introduce the concept of ‘machine intelligence’ and identify successful, real-world applications of such technologies and potential future applications. [SM6m, EA5m]
  2. Explain the theoretical principles of Artificial Neural Networks and Fuzzy Logic , and relate those principles to neural-fuzzy modelling and fuzzy control. [SM6m]
  3. Discriminate between real-world systems that require machine-learning based approaches and those that are more suited for conventional solutions, with a particular emphasis on design and analysis for ill-defined, non-linear, high-dimensional and uncertain systems. [EA3m, EA4m, EP8m]
  4. Provide pragmatic/effective solutions and systematic analysis to complex problems which cannot be necessarily defined mathematically, demonstrating the critically evaluate such solutions over conventional alternatives in terms of performance, ease and costs of implementation, transparency to the user, and ease of future maintenance. [EA3m, EA4m, D1m]
  5. Work with hardware and software to strike the often difficult balance between the theoretical/practical analysis/design of complex ill-defined systems characterised by uncertainties and critically evaluate machine-learning based solutions. [EA5m, EP3m, D1m, EP8m]

This module satisfies the AHEP3 (Accreditation of Higher Education Programmes, Third Edition) Learning Outcomes that are listed in brackets after each learning outcome above. For further details on AHEP3 Learning Outcomes, see the downloads section of this webpage.

Syllabus

  • How can fuzzy systems deal with uncertainty?
  • Fuzzy Sets and their properties
  • Comparison with Classical Sets
  • Operations on Fuzzy Sets
  • Membership Functions
  • Universe of Discourse
  • Fuzzification
  • Defuzzification
  • Fuzzy Rule of Inference
  • Fuzzy Rules Formulation
  • Fuzzy Control Design
  • Neural Fuzzy Modelling
  • Self-Organising Fuzzy Logic Control
  • Fuzzy Predictive Modelling
  • Fuzzy Clustering

Learning and Teaching Methods

Lectures and Tutorials: 24 hours
Laboratories: 12 hours
Independent Study: 64 hours

Learning and Teaching Materials

All teaching materials will be available via MOLE.

Assessment

2 hour examination

Feedback

Although this module does not include marked assignments, it does however involves interactive sessions between the Module Leader (MM) and all students as follows:

  • During 3 "Interactive" Laboratory Sessions dealing with specific problems of modelling control and system identification using Fuzzy technology. MM supervises such sessions personally with the help of able and experienced demonstrators where he can gauge the level of understanding of all students and give them opportunities to gain feedback on their progress.
  • Tutorial Sessions are organised whereby more interactions with students are involved which will enable the module leader to gauge even more the level of engagement and understanding of all students. Students can also ask questions and gain feedback this way.
  • A brief summary of how the class performed in the examination, both on the paper overall and on individual questions will be provided shortly after the examination board. Students will also have the opportunity to view the example answers and to see their scripts.

Student Evaluation

Students are encouraged to provide feedback during the module direct to the lecturer. Students will also have the opportunity to provide formal feedback via the Faculty of Engineering Student Evaluation Survey at the end of the module.

Recommended Reading

  • Passino, K.M. and S. Yurkovich. Fuzzy Control. Addison-Wesley Publishing Company, 1998, ISBN 0-201-18074-X, 2001 [Available in Information Commons, 629.89 (P)]