Introduction to Predictive control for beginners
A useful starting point for any book is to motivate the reader: why do I need this book or topic? Consequently this web book will begin by demonstrating the numerous scenarios where classical control approaches are inadequate. Not too much time is spent on this as this is motivational, but hopefully enough for the reader to be convinced that there is a significant number of real control problems where better approaches are required.
Having given some motivation, we present some insight into how humans deal with such challenging control problems and demonstrate the underlying concepts which ultimately form the main building blocks of predictive control.
The book is organised into a number of brief chapters highlighted below. These resources are intended for beginners rather than advanced users and thus focus on basic concepts and algebra rather than the state of the art in research. The intention is not to go into too much fine detail with respect to recent developments, but instead to concentrate on core concepts, supported by a presentation of the main mathematical development. Some MATLAB files are provided as back up so viewers can follow up any aspects they find puzzling or interesting.
A key message students should take on board is that predictive control represents a way of thinking or approaching control problems, NOT a specific algorithm. It is rarely wise to take an off-the-shelf algorithm as one will only get the most out of MPC by some tailoring to the specific application of interest. Moreover, it is very easy to do a 'bad' MPC design if one is ignorant of the underlying principles.
- Classical control and weaknesses
- Predictive Functional Control
- Finite horizon predictive control laws.
- Infinite horizon predictive control laws.
- Constraint handling.
- Using feedforward information.
Viewers are encouraged to reproduce core results using MATLAB and pen and paper to test their own progress and understanding. It is implicit in these chapters that students have core competence in control concepts and scenarios. Readers can obtain (from the googlesite) some source files, notably MATLAB code, so that they can reproduce results quickly and easily.