Model Predictive Control Made Easy

Workshop on model predictive control for mechatronic systems using Impact

February 27, 2025 @ KU Leuven, Mechanical Engineering, Room De Groote, Leuven

Registration Important dates Organizers Contact

Program Venue Relevant repositories

Overview

In this workshop, participants will engage in hands-on exploration of model predictive control applied to mechatronic systems. By engaging with cutting-edge tools and techniques, participants will develop the skills necessary to configure and deploy model predictive controllers on real hardware.

To streamline the guided exercises, the workshop makes use of the free and open-source Rockit [1] and Impact [2][3] software frameworks developed by the MECO Research Team at KU Leuven and built on top of the numerical optimization framework CasADi [4], designed for efficient nonlinear programming.

Exercises will be mainly in Python and Matlab. Attendees can later adopt the presented open-source software frameworks in their research.

While foundational concepts of nonlinear programming, optimal control and model predictive control will be briefly introduced, the course focuses on learning-by-doing. The course prioritizes practical know-how, enabling participants to directly apply Impact to tackle real-world control challenges.

This workshop is organized by members of the MECO Research Team of core lab MPRO, Flanders Make@KU Leuven, in collaboration with core lab MotionS, Flanders Make. The MECO Research Team focusses on modeling, estimation, identification, analysis and optimal control of motion and motion systems such as mechatronic systems or machine tools. It combines theoretical contributions (development of design methodologies) with experimental knowhow (implementation and experimental validation on lab-scale as well as industrial setups). The theoretical research benefits from the group’s expertise on numerical optimization, especially convex optimization.

The following videos show previous works developed by the MECO Research Team using the software tools that will be used in this workshop:

Registration

Due to capacity constraints at the venue of the workshop, we kindly ask all participants to register their participation.

Participation at the workshop is free of charge, but registration is compulsory. Please contact the organizers in case you have any question.

Lunch will be provided during the workshop.

Use the following button to register. Registration will open in January 2025.

Click here to register for the workshop

Your registration will be completed only after you receive a confirmation email from the organizers.

Important dates

Organizers

This workshop is organized (and its content has been created) by:

Alvaro Florez
Doctoral researcher

Branimir Mrak
Senior Research Engineer

David Kiessling
Doctoral researcher

Jan Swevers
Professor

Joris Gillis
Research expert

Wilm Decré
Research manager

Contact

You can reach the organizers for any questions by contacting them at:

wilm.decre < at > kuleuven.be

Relevant repositories

Workshop's GitHub CasADi Rockit Impact

Program

Tentative.

  • 12:00–13:00
    Lunch

  • 13:00–13:30
    Introduction to nonlinear programming, optimal control and model predictive control

    Preliminary concepts on specifying and solving nonlinear programming problems

  • 13:30–15:00
    Tutorial on Impact

    How to easily specify, prototype and deploy model predictive controllers

  • 15:00–15:15
    Break

  • 15:15–15:45
    NN-MPC

    Neural Network approximated MPC

  • 15:45–16:00
    Concluding remarks

Legend
GENERAL THEORETICAL PRACTICAL

Venue

KU Leuven, Department of Mechanical Engineering, Room De Groote, Leuven

References

[1] J. Gillis, B. Vandewal, G. Pipeleers, J. Swevers. 'Effortless modeling of optimal control problems with Rockit.' 39th Benelux Meeting on Systems and Control 2020, Elspeet, The Netherlands. https://gitlab.kuleuven.be/meco-software/rockit

[2] A. Florez, A. Astudillo, W. Decré, J. Swevers, and J. Gillis, 'IMPACT: A Toolchain for Nonlinear Model Predictive Control Specification, Prototyping, and Deployment', IFAC-PapersOnLine, vol. 56, no. 2, pp. 3164-3169, 2023. https://gitlab.kuleuven.be/meco-software/impact

[3] A. Astudillo, A. Florez, W. Decré, and J. Swevers, 'Rapid Deployment of Model Predictive Control for Robotic Systems: From IMPACT to ROS 2 Through Code Generation', in Proceedings of the 2024 IEEE 18th International Conference on Advanced Motion Control (AMC), 2024.

[4] J. Andersson, J. Gillis, G. Horn, J. Rawlings, and M. Diehl, 'CasADi: a software framework for nonlinear optimization and optimal control', Mathematical Programming Computation, vol. 11, 2019.