Leveraging Reinforcement Learning and Koopman Theory for Enhanced Model Predictive Control Performance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Dony, Md Nur-A-Adam
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913850587086848
author Dony, Md Nur-A-Adam
author_facet Dony, Md Nur-A-Adam
contents This study presents an innovative approach to Model Predictive Control (MPC) by leveraging the powerful combination of Koopman theory and Deep Reinforcement Learning (DRL). By transforming nonlinear dynamical systems into a higher-dimensional linear regime, the Koopman operator facilitates the linear treatment of nonlinear behaviors, paving the way for more efficient control strategies. Our methodology harnesses the predictive prowess of Koopman-based models alongside the optimization capabilities of DRL, particularly using the Proximal Policy Optimization (PPO) algorithm, to enhance the controller's performance. The resulting end-to-end learning framework refines the predictive control policies to cater to specific operational tasks, optimizing both performance and economic efficiency. We validate our approach through rigorous NMPC and eNMPC case studies, demonstrating that the Koopman-RL controller outperforms traditional controllers by achieving higher stability, superior constraint satisfaction, and significant cost savings. The findings indicate that our model can be a robust tool for complex control tasks, offering valuable insights into future applications of RL in MPC.
format Preprint
id arxiv_https___arxiv_org_abs_2505_08122
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Leveraging Reinforcement Learning and Koopman Theory for Enhanced Model Predictive Control Performance
Dony, Md Nur-A-Adam
Systems and Control
93A30 (Mathematical Systems Theory), 68T05 (Learning and Adaptive Systems), 68Q32 (Computational Learning Theory)
This study presents an innovative approach to Model Predictive Control (MPC) by leveraging the powerful combination of Koopman theory and Deep Reinforcement Learning (DRL). By transforming nonlinear dynamical systems into a higher-dimensional linear regime, the Koopman operator facilitates the linear treatment of nonlinear behaviors, paving the way for more efficient control strategies. Our methodology harnesses the predictive prowess of Koopman-based models alongside the optimization capabilities of DRL, particularly using the Proximal Policy Optimization (PPO) algorithm, to enhance the controller's performance. The resulting end-to-end learning framework refines the predictive control policies to cater to specific operational tasks, optimizing both performance and economic efficiency. We validate our approach through rigorous NMPC and eNMPC case studies, demonstrating that the Koopman-RL controller outperforms traditional controllers by achieving higher stability, superior constraint satisfaction, and significant cost savings. The findings indicate that our model can be a robust tool for complex control tasks, offering valuable insights into future applications of RL in MPC.
title Leveraging Reinforcement Learning and Koopman Theory for Enhanced Model Predictive Control Performance
topic Systems and Control
93A30 (Mathematical Systems Theory), 68T05 (Learning and Adaptive Systems), 68Q32 (Computational Learning Theory)
url https://arxiv.org/abs/2505.08122