ES915 Synthetic Intelligence and Intelligent Software Design (12 CATS) |
|
Lecturers: Dr E L Hines, BSc, PhD, GDMA, AMIEE Professor P J Bryanston-Cross, BSc, PhD, CEng, FRAeS |
|
| The area of what is sometimes called ‘Synthetic Intelligence’ is developing very rapidly. It is concerned with a number of subject areas including Artificial Neural Networks, Fuzzy Systems (or Fuzzy Logic), Genetic Algorithms (or Evolutionary Systems) and others. As such they are allied in a very general sense to what one might call Artificial Intelligence. This may also be referred to as Soft Computing. These are also related to the ideas of Data Fusion, Data Mining, Data Warehousing, Data Visualisation, Intelligent Agents, and others. The aim of this module is to put these ideas into context by looking at some applications of this approach to the solution of some engineering problems. Lecture material will be supported by the use of for example the Matlab environment to explore simulation exercises based on sensor data. A number of videos will be used to give a snap shot of some of the international applications and research work, which is being undertaken in these areas. The course will provide students with ‘hands on’ experience through coursework exercises, which involves the application of these techniques to the analysis of sensor data. The data may be derived from any number of sources for example imaging (e.g. particle image velocimetry), medical (e.g. brain scans, ophthalmic), non-destructive testing images of composite material, spectral imaging of the combustion process in instrumented intelligent engines. Another type of data, which we shall explore, is derived from our Electronic Nose (also called Artificial Nose, e-nose, etc). | |
Aims :
This module aims to:
|
|||||||||||||||||||||
Objectives:
At completion, students will have:
|
|||||||||||||||||||||
| Syllabus : Historical perspective. Information processing from statistics, expert systems and neural networks. How to estimate performance and which technique is best. Introduction to fundamentals of Neural Networks (typical paradigms including: Perceptron, Multilayer Perceptron, and Kohonen networks); Fuzzy Logic; Neuro-Fuzzy systems; Natural algorithms (Darwinian filtering, Genetic algorithms, Simulated annealing, Chaos, Fractals, Evolutionary algorithms). System design and implementation considerations (eg. Hardware versus software, Network topology, Growing vs pruning, Genetic Algorithms, Evolutionary programming). Introduction to Intelligent Agents; Concept of Conscious systems; Data mining and Data fusion. Applications case studies eg: Intelligent sensors, Image processing, Character recognition, Instrumentation, Control, etc. Commercial perspective. | |||||||||||||||||||||
| Illustrative
Bibliography: R R Brooks, S Iyengar, Multi-Sensor Fusion: Fundamentals and Applications With Software, Prentice Hall (1997) R Cotterill, Enchanted Looms: Conscious networks in brain and computers, Cambridge University Press (1998). S M Weiss and N Indurkhya, Predictive data mining, Morgan Kaufmann Publishers (1997). T J Ross, Fuzzy Logic with Engineering Applications, McGraw-Hill (1995) C-H Lin and C S G Lee, Neural Fuzzy Systems: A Neuro-Fuzzy Synergism to Intelligent Systems, Prentice-Hall (1996). S M Weiss and C A Kulikowski, Computer systems that learn, Morgan Kaufmann, 1991. M Negnevitsky, Artificial intelligence: A guide to intelligent systems, Addison Wesley (2002). A A Hopgood, Intelligent Systems for Engineers and Scientists, CRC Press (2001). S K Pal and S Mitra, Neuro-fuzzy pattern recognition: Methods in soft computing, Wiley (1999). J P Bigus and J Bigus, Constructing intelligent agents with Java, Wiley (2001). |
|||||||||||||||||||||
Assessment:
|
|||||||||||||||||||||