Intelligent Systems and Control

Overview

Intelligent Systems and Control (ELE8066) develops a robust understanding of the major academic topics which define control methods and intelligent algorithms in dynamic systems. The course contents includes:
Semester 1 (control):
• Software simulation of dynamical systems
• State-space modelling and analysis of dynamical systems
•Control design methods in state-space
• State observer design
• Stability analysis
• Introduction to advanced control methods

Semester 2 (Intelligent Systems):
• Introduction to artificial intelligence
• Neural networks and training algorithms
• Metaheuristic Methods (Genetic Algorithms, Particle Swarm Optimisation)
• Fuzzy logic and fuzzy control systems

The module has a final written examination and two individual coursework elements (one in each semester), that are a combination of design calculations, theoretical derivations, algorithm development and practical work in Matlab. Each coursework accounts for 20% of the final mark, while the final exam contributes 60%.

Semester 1 is focused on understanding control problem specifications and objectives, being able to model, simulate and formally analyse the dynamic behaviour of a system using state space methods and being able to design state feedback controllers such that the closed-loop system meets desired performance objectives. In semester 2 the focus is on gaining an understanding of the basics of a range of intelligent systems techniques, and how to apply them to solve practical engineering problems.

Learning Objectives

• Apply analytical methods to derive dynamic models
• Analyse system properties in state space (e.g., controllability, observability, stability etc)
• Convert from state-space (SS) to linear transfer function (LTF) representation and vice versa
• Develop stabilising state feedback controllers for linear dynamical systems
• Develop state observers for linear dynamical systems
• Apply Lyapunov methods to assess the stability of linear and nonlinear dynamical systems
• Understand and apply simple instances of advanced control methods
• Understand the basic principles of a range of intelligent systems techniques methods
• Design neural network models for classification and modelling tasks
• Describe the training methodology and the design choices that apply when training neural networks
• Design metaheuristic algorithms for practical optimisation problems
• Develop fuzzy control laws for practical applications
• Use software for simulation, control and estimation.
• Use software to apply computational techniques (e.g. Neural Networks, GAs) to solve a practical problem.

Skills

The module will give you experience of how to model, simulate and analyse linear and non-linear systems using software tools (e.g., Matlab), knowledge of linear state-space methods and how to apply them to model, analyse and design control laws for dynamical systems, and knowledge of a range of intelligent systems techniques and how to apply them to solve practical engineering problems.

Assessment

None