20810400 - ADVANCED CONTROL SYSTEMS

The course provides the skills and knowledge for the advanced control of complex systems, which may consist of multiple agents, such as multiple robotic platforms. It will present distributed control methodologies for multi-agent systems, where each robot determines its control actions based on local information. To this end, the course will cover the fundamentals of algebraic graph theory and their application to coordination problems, such as consensus, formation control, and distributed optimization.
The topic of predictive control will also be addressed, starting from the foundations of optimal control. The course will introduce the basics of formulating and solving Model Predictive Control (MPC) problems applied to complex, multivariable dynamic processes. Furthermore, the main MPC formulations will be analyzed, including centralized and distributed approaches, with a particular focus on systems composed of multiple interacting subsystems subject to constraints.
teacher profile | teaching materials

Programme

Introduction to multi-agent systems

Graph theory and algebraic graph theory

Consensus protocols with different graphs

Formation control

Control of robot swarms and further coordination problems

Introduction to Model Predictive Control and mathematical fundamentals

Formulation of the MPC Problem

Stability of MPC

MPC for Multi-agent Systems

Prerequisiti

Fundamentals Of Automatic Control and possibly Dynamics and Control of Complex Systems and Robotics

Core Documentation

- F. Bullo, Lectures on Network Systems, CreateSpace Independent Publishing Platform, ISBN ‎978-1986425643, 2022

- Camacho, E. F., & Bordons, C., Model Predictive control, Springer, ISBN: 978-1-85233-694-3, 2013.

- Lecture notes

Reference Bibliography

- Veysel Gazi and Kevin Passino. 2011. Swarm Stability and Optimization. Springer Publishing Company, Incorporated. ISBN 978-3642422706 - F. Bullo, J. Cortes, and S. Martinez, Distributed Control of Robotic Networks: A Mathematical Approach to Motion Coordination Algorithms, Princeton University Press, Princeton, NJ, USA, 2009. - M. Mesbahi and M. Egerstedt, Graph Theoretic Methods for Multiagent Networks, Princeton University Press, Princeton, NJ, 2010. - Maciejowski, J., Predictive Control with Constraints, Pearson Education POD, 2002. - Rossiter, J. A., Model-Based Predictive Control: A Practical Approach, CRC Press, 2003. - Maestre, J. M., & Negenborn, R. R. (Eds.). (2014). Distributed model predictive control made easy (Vol. 69). Dordrecht, Netherlands: Springer.

Attendance

Attendance is strongly recommended given the practical nature of the course.

Type of evaluation

The exam consists of an oral test to discuss the theoretical aspects of the course and a project to explore the practical aspects of the course.

teacher profile | teaching materials

Programme

Multi-agent systems:
- Introduction to multi-agent systems: examples and distributed coordination problems.
- Graph theory and algebraic graph theory: directed and undirected graphs, connectivity, the Laplacian matrix and its spectral properties.
- Consensus protocols: convergence on fixed, directed and time-varying graphs.
- Formation control: displacement-based and distance-based formations.
- Robot swarms: potential-based coordination
Model Predictive Control:
- Introduction to Model Predictive Control and mathematical fundamentals
- Formulation of the MPC Problem
- Stability of MPC
- MPC for Multi-agent Systems


Prerequisiti

Fundamentals Of Automatic Control and possibly Dynamics and Control of Complex Systems and Robotics

Core Documentation

- M. Mesbahi and M. Egerstedt, Graph Theoretic Methods for Multiagent Networks, Princeton University Press, Princeton, NJ, 2010.
- Camacho, E. F., & Bordons, C., Model Predictive control, Springer, ISBN: 978-1-85233-694-3, 2013.
- Lecture notes


Reference Bibliography

- Veysel Gazi and Kevin Passino. 2011. Swarm Stability and Optimization. Springer Publishing Company, Incorporated. ISBN 978-3642422706 - F. Bullo, J. Cortes, and S. Martinez, Distributed Control of Robotic Networks: A Mathematical Approach to Motion Coordination Algorithms, Princeton University Press, Princeton, NJ, USA, 2009. - F. Bullo, J. Cortes, and S. Martinez, Distributed Control of Robotic Networks: A Mathematical Approach to Motion Coordination Algorithms, Princeton University Press, Princeton, NJ, USA, 2009. - Maciejowski, J., Predictive Control with Constraints, Pearson Education POD, 2002. - Rossiter, J. A., Model-Based Predictive Control: A Practical Approach, CRC Press, 2003. - Maestre, J. M., & Negenborn, R. R. (Eds.). (2014). Distributed model predictive control made easy (Vol. 69). Dordrecht, Netherlands: Springer.

Attendance

Attendance is strongly recommended given the practical nature of the course.

Type of evaluation

The exam consists of an oral test to discuss the theoretical aspects of the course and a project to explore the practical aspects of the course. Students are assessed according to the Dublin descriptors, taking into account their level of knowledge, understanding and critical elaboration of the topics covered in the course. Students who show that they have acquired the fundamental concepts and can apply them in simple contexts will receive a pass mark. Average marks will be given to students who, besides mastering the basics, show the ability to analyse and independently rework more complex concepts. The highest marks are reserved for students who show in-depth understanding, the ability to synthesize, critical thinking, and independence in solving problems, including new problems or problems not explicitly addressed during the course.