Nonlinear Control of Uncertain Systems

Conventional and Learning-Based Alternatives with Python
Erscheinungsjahr 2026
ISBN: 978-3-031-93286-1
Verlag: Springer

Buch, Englisch, 623 Seiten, Format (B × H): 160 mm x 241 mm, Gewicht: 1125 g

This book provides scalable, effective and real-world-compatible methods and algorithms for the control and extended estimation of uncertain nonlinear systems against a backdrop of often-unconventional problems. The author provides advice on choosing which solution is most relevant to the desired control objectives, the nature of present uncertainties and the impact on closed-loop performance.

The book introduces its key paradigms step by step and then presents the family of candidate solutions in detail along with associated python scripts. It helps the reader develop a critical and comparative point of view and thus to distinguish the best choice of solutions, some of which prove to be conventional and others to employ advanced learning-based methods. This book shows how each category applies to specific groups of problems, but the choice is made based on pragmatic assessments of efficiency and efficacy rather than on dogmatic adherence to the benefits of one or the other.

All of the concepts and solutions described in the text are illustrated using significantly challenging problems, wherever possible with real-world relevance. Solutions are implemented using Python scripts, freely downloadable from the author’s GitHub account. Practical features such as messages, cautions, summaries and important comments are clearly presented to aid reading, retention and recall.

appeals to both academics and professional practitioners studying and developing nonlinear industrial control systems; its critical comparative appraisal and detailed range of solutions help readers to navigate a complex taxonomy of systems and to find the right solution—learning-based or conventional—for the problems before them.

Zielgruppe

Professional/practitioner

Autoren/Hrsg.

Alamir, Mazen

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Maschinelles Lernen
• Mathematik | Informatik EDV | Informatik Daten / Datenbanken
• Technische Wissenschaften Verkehrstechnik | Transportgewerbe Fahrzeugtechnik
• Technische Wissenschaften Elektronik | Nachrichtentechnik Nachrichten- und Kommunikationstechnik Regelungstechnik
• Interdisziplinäres Wissenschaften Wissenschaften: Forschung und Information Kybernetik, Systemtheorie, Komplexe Systeme

Inhaltsverzeichnis

Part I: Definitions, Notation, Concepts and Tools.- What is this Book About?.- Definitions, Notation and Main Concepts.- Quick Python Reminders and Key Modules.- Part II: Methodologies and Algorithms.- Handling Uncertainties via Standard Methods.- Solving Deterministic Model Predictive Control Problems.- A Framework and a Python-Package for Real-Time Nonlinear Model Predictive Control Parameters Settings.- Designing an Uncertainty-Aware Dynamic Output Feedback via Deterministic Optimal Control Solutions.- Nonlinear Moving Horizon Extended Observers.- Further Advanced Topics.- Control of Automotive Automated Manual Transmission.- Power Management in an EV-Charging Station.- Feedback Law with Probabilistic Certification for Propofol-Based Control of BIS During Anesthesia.- Learning Optimal Energy Management in Hybrid Vehicles.- Q-Learning Solution to the Combined Therapy of Cancer.- Decentralized Frequency Control in Micro-grids.

Über Autor(innen)

Mazen Alamir is Research Director at CNRS, France. He graduated in mechanics (Grenoble, 1990) and avionics (Toulouse 1992). He received his Ph.D. in nonlinear model predictive control in 1995 from Grenoble Institute of Technology. He served as Head of the nonlinear systems and complexity research group at the Control System Department, University of Grenoble-Alpes. His main research topics are nonlinear model predictive control, nonlinear moving-horizon estimators and blind anomaly detection in engineering equipments. He was Member of the IFAC technical committee on nonlinear systems as well as the IEEE Conference Editorial Board and served as Associate Editor of the (2008-2021). He organized the first IFAC workshop on NMPC for Fast Systems, Grenoble 2006. He is Co-Founder and Scientific Advisor of the startup Amiral-Technologies, specialized in AI algorithms for industrial predictive maintenance and diagnosis, and Winner of the 2017 GE-digital industrial challenge.