MSc Advanced Control and Systems Engineering / Course details

The unit has two different parts.

Part A - System Identification

• Examplar system identification problems.
• Measurements and statistics.
• Non-parametric methods: Time and frequency domain
• Least square problem. Statistic foundation.
• Parametric methods (ARX, OE).
• Input design.
• Optimisation: gradient method for OE.
• Recursive estimation.
• Validation.

Part B - Artificial Intelligence (for dynamic systems modelling)

• Introduction to AI.
• Neural network models (single-layer and multiple-layer neural networks).
• Deep learning (neural networks).
• Learning and optimisation methods.
• Performance evaluation.

The unit aims to give students an understanding of how system identification and artificial intelligent algorithms can be used to find models of dynamic systems; how least squares and optimisation approaches can be used for parameter estimation; the influence of noise on the parameter estimation and the relevance of measurement theory for the identification process.

On successful completion of the course, a student will be able to:

ILO 1: Test and validate the neural networks models with different selecting criteria.

ILO 2: Optimise the neural networks model parameters with input-output data.

ILO 3: Demonstrate understanding of the potential and limitations of AI and their essential steps for dynamic system modelling, including choices of AI models and performance evaluation.

ILO 4: Describe the implementation of the data-driven techniques that can be used to identify linear and nonlinear dynamic systems.

ILO 5: Use statistical analysis techniques to describe random variation in measured data.

The unit has two different parts.

Part A - System Identification

• Examplar system identification problems.
• Measurements and statistics.
• Non-parametric methods: Time and frequency domain
• Least square problem. Statistic foundation.
• Parametric methods (ARX, OE).
• Input design.
• Optimisation: gradient method for OE.
• Recursive estimation.
• Validation.

Part B - Artificial Intelligence (for dynamic systems modelling)

• Introduction to AI.
• Neural network models (single-layer and multiple-layer neural networks).
• Deep learning (neural networks).
• Learning and optimisation methods.
• Performance evaluation.

Lectures, tutorial and laboratory.

Examination - feedback will be given after the exam board.

System Identifaction coursework - individual feedback is provided 3 weeks after submission.

AI coursework - individual feedback is provided 3 weeks after submission.

Söderström, T. (1989). System Identification. Prentice Hall.
Ljung, L. (1987). System Identification: Theory for the User. Prentice-Hall.
Goodwin, G.C., & Payne, R.L. (Ed.). (1977). Dynamic System Identification: Experiment Design and Data Analysis. Academic Press.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
Alpaydin, E. (2020). Introduction to Machine Learning (4th ed.). MIT Press