Approximating MPC Solutions Using Deep Neural Networks: Towards Application in Mechatronic Systems

Edward Kikken, Jeroen Willems, Branimir Mrak, Bruno Depraetere

2025

Abstract

Model Predictive Control is an advanced control technique that can yield high performance, but it is often challenging to implement. Especially for systems with dynamics that are complex to model, have strong nonlinearities, and/or have small time constants, it is often not possible to complete the needed online optimizations fast and reliable enough. In this work we look at approximating the MPC solutions using black-box models i.e. deep neural networks, so that the computational load at runtime is strongly reduced. We use a supervised learning approach to train these models to yield outputs similar to those of an example dataset of offline pre-computed MPC solutions. We illustrate this approach on three realistic (active-suspension system, parallel robot, and a truck-trailer), illustrating the typical workflow and how the approach has to be set up to address the varying challenges. We show that the approximate MPC solutions yield a high level of performance, reaching nearly the level of the original MPC, yet at a strongly reduced computational load.

Paper Citation

in Harvard Style

Kikken E., Willems J., Mrak B. and Depraetere B. (2025). Approximating MPC Solutions Using Deep Neural Networks: Towards Application in Mechatronic Systems. In Proceedings of the 22nd International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO; ISBN 978-989-758-770-2, SciTePress, pages 71-81. DOI: 10.5220/0013712900003982

in Bibtex Style

@conference{icinco25,
author={Edward Kikken and Jeroen Willems and Branimir Mrak and Bruno Depraetere},
title={Approximating MPC Solutions Using Deep Neural Networks: Towards Application in Mechatronic Systems},
booktitle={Proceedings of the 22nd International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO},
year={2025},
pages={71-81},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013712900003982},
isbn={978-989-758-770-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 22nd International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO
TI - Approximating MPC Solutions Using Deep Neural Networks: Towards Application in Mechatronic Systems
SN - 978-989-758-770-2
AU - Kikken E.
AU - Willems J.
AU - Mrak B.
AU - Depraetere B.
PY - 2025
SP - 71
EP - 81
DO - 10.5220/0013712900003982
PB - SciTePress