Reinforcement learning-based model predictive control for greenhouse climate control

Greenhouse climate control is concerned with maximizing performance in terms of crop yield and resource efficiency. One promising approach is model predictive control (MPC), which leverages a model of the system to optimize the control inputs, while enforcing physical constraints. However, predictio...

Full description

Saved in:
Bibliographic Details
Main Authors: Samuel Mallick, Filippo Airaldi, Azita Dabiri, Congcong Sun, Bart De Schutter
Format: Article
Language:English
Published: Elsevier 2025-03-01
Series:Smart Agricultural Technology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772375524003551
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850054167811325952
author Samuel Mallick
Filippo Airaldi
Azita Dabiri
Congcong Sun
Bart De Schutter
author_facet Samuel Mallick
Filippo Airaldi
Azita Dabiri
Congcong Sun
Bart De Schutter
author_sort Samuel Mallick
collection DOAJ
description Greenhouse climate control is concerned with maximizing performance in terms of crop yield and resource efficiency. One promising approach is model predictive control (MPC), which leverages a model of the system to optimize the control inputs, while enforcing physical constraints. However, prediction models for greenhouse systems are inherently inaccurate due to the complexity of the real system and the uncertainty in predicted weather profiles. For model-based control approaches such as MPC, this can degrade performance and lead to constraint violations. Existing approaches address uncertainty in the prediction model with robust or stochastic MPC methodology; however, these necessarily reduce crop yield due to conservatism and often bear higher computational loads. In contrast, learning-based control approaches, such as reinforcement learning (RL), can handle uncertainty naturally by leveraging data to improve performance. This work proposes an MPC-based RL control framework to optimize the climate control performance in the presence of prediction uncertainty. The approach employs a parametrized MPC scheme that learns directly from data, in an online fashion, the parametrization of the constraints, prediction model, and optimization cost that minimizes constraint violations and maximizes climate control performance. Simulations show that the approach can learn an MPC controller that significantly outperforms the current state-of-the-art in terms of constraint violations and efficient crop growth.
format Article
id doaj-art-c416f1f4c7fb4de4a18ec6c8f13af9e3
institution DOAJ
issn 2772-3755
language English
publishDate 2025-03-01
publisher Elsevier
record_format Article
series Smart Agricultural Technology
spelling doaj-art-c416f1f4c7fb4de4a18ec6c8f13af9e32025-08-20T02:52:20ZengElsevierSmart Agricultural Technology2772-37552025-03-011010075110.1016/j.atech.2024.100751Reinforcement learning-based model predictive control for greenhouse climate controlSamuel Mallick0Filippo Airaldi1Azita Dabiri2Congcong Sun3Bart De Schutter4Delft Center for Systems and Control, Delft University of Technology, 2628 CD, Delft, the Netherlands; Corresponding author.Delft Center for Systems and Control, Delft University of Technology, 2628 CD, Delft, the NetherlandsDelft Center for Systems and Control, Delft University of Technology, 2628 CD, Delft, the NetherlandsAgricultural Biosystems Engineering Group, Wageningen University, 6700 AA, Wageningen, the NetherlandsDelft Center for Systems and Control, Delft University of Technology, 2628 CD, Delft, the NetherlandsGreenhouse climate control is concerned with maximizing performance in terms of crop yield and resource efficiency. One promising approach is model predictive control (MPC), which leverages a model of the system to optimize the control inputs, while enforcing physical constraints. However, prediction models for greenhouse systems are inherently inaccurate due to the complexity of the real system and the uncertainty in predicted weather profiles. For model-based control approaches such as MPC, this can degrade performance and lead to constraint violations. Existing approaches address uncertainty in the prediction model with robust or stochastic MPC methodology; however, these necessarily reduce crop yield due to conservatism and often bear higher computational loads. In contrast, learning-based control approaches, such as reinforcement learning (RL), can handle uncertainty naturally by leveraging data to improve performance. This work proposes an MPC-based RL control framework to optimize the climate control performance in the presence of prediction uncertainty. The approach employs a parametrized MPC scheme that learns directly from data, in an online fashion, the parametrization of the constraints, prediction model, and optimization cost that minimizes constraint violations and maximizes climate control performance. Simulations show that the approach can learn an MPC controller that significantly outperforms the current state-of-the-art in terms of constraint violations and efficient crop growth.http://www.sciencedirect.com/science/article/pii/S2772375524003551Greenhouse climate controlModel predictive controlReinforcement learning
spellingShingle Samuel Mallick
Filippo Airaldi
Azita Dabiri
Congcong Sun
Bart De Schutter
Reinforcement learning-based model predictive control for greenhouse climate control
Smart Agricultural Technology
Greenhouse climate control
Model predictive control
Reinforcement learning
title Reinforcement learning-based model predictive control for greenhouse climate control
title_full Reinforcement learning-based model predictive control for greenhouse climate control
title_fullStr Reinforcement learning-based model predictive control for greenhouse climate control
title_full_unstemmed Reinforcement learning-based model predictive control for greenhouse climate control
title_short Reinforcement learning-based model predictive control for greenhouse climate control
title_sort reinforcement learning based model predictive control for greenhouse climate control
topic Greenhouse climate control
Model predictive control
Reinforcement learning
url http://www.sciencedirect.com/science/article/pii/S2772375524003551
work_keys_str_mv AT samuelmallick reinforcementlearningbasedmodelpredictivecontrolforgreenhouseclimatecontrol
AT filippoairaldi reinforcementlearningbasedmodelpredictivecontrolforgreenhouseclimatecontrol
AT azitadabiri reinforcementlearningbasedmodelpredictivecontrolforgreenhouseclimatecontrol
AT congcongsun reinforcementlearningbasedmodelpredictivecontrolforgreenhouseclimatecontrol
AT bartdeschutter reinforcementlearningbasedmodelpredictivecontrolforgreenhouseclimatecontrol