Intelligent Control and Signal Processing from PELINCEC Perspective

Presented by Bartłomiej Beliczyński

Pelincec Internal Seminar, Warsaw December 28, 2004, „Intelligent Control and Signal Processing” Group.

Scope of the Presentation

{ { { { { { {

Comments on terminology and meaning Intelligent control and signal processing elsewhere WG3 research topics Working methods The implementation: seminars and workshops Future Conclusions

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control meaning – the first attempt and the task force

{

{

For many, the term intelligent control has come to mean some form of control using fuzzy and/or neural network methodologies. Panos Antsaklis’ Task Force z

„The task force has not attempted to address the issue of intelligence in its generality, but instead has concentrated on deriving working characterizations of Intelligent Control”

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control Versus Conventional Control {

{

{

{

The word „control” in „intelligent control” has different - more general meaning than the word „control” in „conventional control”. The area of intelligent control is interdisciplinary and it attempts to combine theories and methods from areas such as control, computer science and operations research. Intelligent controllers are emulating human mental faculties such as adaptation and learning dealing with uncertainty and coping with large amounts of data. An alternative term is “autonomous (intelligent) control”.

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Control and Intelligent Systems {

{

{

{

An intelligent system must define and use goals Control system is required to move the system towards these goals Thus any intelligent system will be (or consists of as a part) control system, „Intelligent Control Systems”, „Intelligent Systems”

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent system properties {

{

{

Minimum: ability to sense environment, to make decisions and control Higher: ability to recognize objects and events, to represent knowledge, to reason and plan Advanced: ability to perceive and understand and to choose visely

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

A Working Definition „An intelligent system must be highly adaptable to significant unanticipated changes, and so learning is essential. It must exhibit high degree of autonomy in coping with changes. It must be able to deal with significant complexity, and this leads to certain sparse types of functional architectures such as hierarchies.” P.Antsaklis’ Task Force (1993). Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Signal Processing „ISP differs fundamentally from the classical approach to statistical signal processing in that the input-output behavior of a complex system is modeled by using “intelligent” or “model-free” techniques, rather than relying on the shortcomings of a mathematical model”. IEEE Press adv. Of S.Haykin and B.Kosko 2001 book „ Intelligent Signal Processing”. Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Samples of Activity in the Field (Elsewhere) {

Pattern Analysis and Machine Intelligence Group Waterloo University, Canada z z z z z

z

Autonomous Systems Machine Perception and Sensors Image Processing and Analysis Pattern Analysis and Recognition Knowledge Systems Focus Groups: Multiple Classifier Systems, Text Mining, Machine Perception and Sensors, Autonomous Systems and Real Time Distributed Design, Shape Analysis, Ultrasound Sensors

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Samples of Activity in the Field (Elsewhere) {

Inteligent Signal Processing-University of Barcelona z z z z z

z z

Central Research interests: Electronic Nose & Artificial Olfaction Taste Sensors & Electronic Tongue Chemical Sensing Technologies Soft Computing for Pattern Recognition and Regression: Artificial Neural Networks Fuzzy Inference Systems

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Samples of Activity in the Field (Elsewhere) {

Inteligent Signal Processing-University of Barcelona cont. z z z z z z

z z

Genetic Algorithms Chemometrics Microcontroller, DSP, PC embedded systems Peripheric research interests: Bio-inspired olfactory signal processing Human Perception: sound, vision, odours, taste, tactile... Multi-modal data fusion Data Mining...

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

WG3 working method and publication life cycle {

{ { { { { {

Preparation (seminars and internal workshops) – problem formulation and solution Title and summary delivery Presenting the paper during external workshop Improving and extending the papers Delivering final version of the papers Reviewing process Publishing

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

The Workshops {

{

Impact of Artificial Intelligence Methods and Models on Engineering Practice – 2003 Bridges Through Time: Intelligent Control, Signal Processing and Real-Time Process Control -2004

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control and Signal Processing Group Research Topics - Fundamentals

z

z

z z z z

The relationship between the singular compartmental systems and the electrical circuits Random sampling for in digital alias-free signal processing Characterisation of Smooth Nonlinear Systems Model structures and properties Realization problems Fractional dynamical systems

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control and Signal Processing Group Topics Research - Control z z z z z

Observer design for uncertain systems Evolutionary methods of trajectory planning Non-linear control Model reference adaptive control Passivity based control

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control and Signal Processing Group Research Topics - Arificial Intelligence Methods

z z z

z z z z z

AI methods in identification and control Cellular Neural Networks Computational intelligence in manufacturing quality control Diagnostics and fault detection Rough Sets and Petri Nets Text categorisation Voice recognition Utilising chaos

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control and Signal Processing Group Research Topics – Mobile Robots

z z z z z

Mobile and social robots LEGO system in the University Lab Mobile robots navigation Route planning and obstactles avoiding Life creatures and walking machines

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control and Signal Processing Group Research Topics – Image Processing

z

z z z

Mathematical morphology for feature extraction in pattern recognition Motion analysis and optical flow computation Old movies processing Gabor filters

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Intelligent Control and Signal Processing Group Topics – Modelling and Simulation

z z z

z

Development of behavioural system modelling UML modelling Modelling of real-time systems with UML language Dicrete-time systems modelling and simulation

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Impact of Artificial Intelligence Methods and Models on Engineering Practice { {

2 full days program, 21 papers all together including z z z

{

2 tutorials 2 invited papers 6 Ph.D. students papers

4 sessions: Image Processing, Approximation and Control, Arificial Intelligence Applications, Robotics

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Cover design

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Graphics’ draft

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Bridges Through Time: Intelligent Control, Signal Processing and Real-Time Process Control

{ {

2 full days program, 26 papers all together including z z

{

3 invited papers 8 Ph.D. students’ papers

4 sessions: Non-linear Systems and Neural Modelling, Automation, Robotics and Software Engineering, Fundamentals and Applications, Image processing

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Conclusion points

{ {

{

{ {

Interdisciplinary field of research Only certain topics are selected and undertaken The near future: undergoing publishing process Pelincec conference preparation Future colaboration

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group

Stairs at the Main Hall of Warsaw University of Technology

Pelincec Internal Seminar, Warsaw December 28, 2004. „Intelligent Control and Signal Processing” Group