A comparative study of deep reinforcement learning for crop production management
Crop production management is essential for optimizing yield and minimizing a field's environmental impact to crop fields, yet it remains challenging due to the complex and stochastic processes involved. Recently, researchers have turned to machine learning to address these complexities. Specif...
| Päätekijät: | , , , , |
|---|---|
| Aineistotyyppi: | Artikkeli |
| Kieli: | English |
| Julkaistu: |
Elsevier
2025-03-01
|
| Sarja: | Smart Agricultural Technology |
| Aiheet: | |
| Linkit: | http://www.sciencedirect.com/science/article/pii/S2772375525000863 |
| _version_ | 1826545670332350464 |
|---|---|
| author | Joseph Balderas Dong Chen Yanbo Huang Li Wang Ren-Cang Li |
| author_facet | Joseph Balderas Dong Chen Yanbo Huang Li Wang Ren-Cang Li |
| author_sort | Joseph Balderas |
| collection | DOAJ |
| description | Crop production management is essential for optimizing yield and minimizing a field's environmental impact to crop fields, yet it remains challenging due to the complex and stochastic processes involved. Recently, researchers have turned to machine learning to address these complexities. Specifically, reinforcement learning (RL), a cutting-edge approach designed to learn optimal decision-making strategies through trial and error in dynamic environments, has emerged as a promising tool for developing adaptive crop management policies. RL models aim to optimize long-term rewards by continuously interacting with the environment, making them well-suited for tackling the uncertainties and variability inherent in crop management. Studies have shown that RL can generate crop management policies that compete with, and even outperform, expert-designed policies within simulation-based crop models. In the gym-DSSAT crop model environment, one of the most widely used simulators for crop management, proximal policy optimization (PPO) and deep Q-networks (DQN) have shown promising results. However, these methods have not yet been systematically evaluated under identical conditions. In this study, we evaluated PPO and DQN against static baseline policies across three different RL tasks, fertilization, irrigation, and mixed management, provided by the gym-DSSAT environment. To ensure a fair comparison, we used consistent default parameters, identical reward functions, and the same environment settings. Our results indicate that PPO outperforms DQN in fertilization and irrigation tasks, while DQN excels in the mixed management task. This comparative analysis provides critical insights into the strengths and limitations of each approach, advancing the development of more effective RL-based crop management strategies. |
| first_indexed | 2025-03-14T05:21:33Z |
| format | Article |
| id | doaj.art-755b975587ee42799dd01ac3326eb6ef |
| institution | Directory Open Access Journal |
| issn | 2772-3755 |
| language | English |
| last_indexed | 2025-03-14T05:21:33Z |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj.art-755b975587ee42799dd01ac3326eb6ef2025-03-06T05:46:49ZengElsevierSmart Agricultural Technology2772-37552025-03-0110100853A comparative study of deep reinforcement learning for crop production managementJoseph Balderas0Dong Chen1Yanbo Huang2Li Wang3Ren-Cang Li4Department of Mathematics, University of Texas at Arlington, Arlington, TX 76019, USADepartment of Agricultural & Biological Engineering, Mississippi State University, Mississippi State, MS 39762, USAUSDA-ARS Genetics and Sustainable Agriculture Research Unit, Mississippi State, MS 39762, USA; Corresponding author.Department of Mathematics, University of Texas at Arlington, Arlington, TX 76019, USADepartment of Mathematics, University of Texas at Arlington, Arlington, TX 76019, USACrop production management is essential for optimizing yield and minimizing a field's environmental impact to crop fields, yet it remains challenging due to the complex and stochastic processes involved. Recently, researchers have turned to machine learning to address these complexities. Specifically, reinforcement learning (RL), a cutting-edge approach designed to learn optimal decision-making strategies through trial and error in dynamic environments, has emerged as a promising tool for developing adaptive crop management policies. RL models aim to optimize long-term rewards by continuously interacting with the environment, making them well-suited for tackling the uncertainties and variability inherent in crop management. Studies have shown that RL can generate crop management policies that compete with, and even outperform, expert-designed policies within simulation-based crop models. In the gym-DSSAT crop model environment, one of the most widely used simulators for crop management, proximal policy optimization (PPO) and deep Q-networks (DQN) have shown promising results. However, these methods have not yet been systematically evaluated under identical conditions. In this study, we evaluated PPO and DQN against static baseline policies across three different RL tasks, fertilization, irrigation, and mixed management, provided by the gym-DSSAT environment. To ensure a fair comparison, we used consistent default parameters, identical reward functions, and the same environment settings. Our results indicate that PPO outperforms DQN in fertilization and irrigation tasks, while DQN excels in the mixed management task. This comparative analysis provides critical insights into the strengths and limitations of each approach, advancing the development of more effective RL-based crop management strategies.http://www.sciencedirect.com/science/article/pii/S2772375525000863Crop production managementReinforcement learningDeep learningMachine learningProximal policy optimizationDeep Q-networks |
| spellingShingle | Joseph Balderas Dong Chen Yanbo Huang Li Wang Ren-Cang Li A comparative study of deep reinforcement learning for crop production management Smart Agricultural Technology Crop production management Reinforcement learning Deep learning Machine learning Proximal policy optimization Deep Q-networks |
| title | A comparative study of deep reinforcement learning for crop production management |
| title_full | A comparative study of deep reinforcement learning for crop production management |
| title_fullStr | A comparative study of deep reinforcement learning for crop production management |
| title_full_unstemmed | A comparative study of deep reinforcement learning for crop production management |
| title_short | A comparative study of deep reinforcement learning for crop production management |
| title_sort | comparative study of deep reinforcement learning for crop production management |
| topic | Crop production management Reinforcement learning Deep learning Machine learning Proximal policy optimization Deep Q-networks |
| url | http://www.sciencedirect.com/science/article/pii/S2772375525000863 |
| work_keys_str_mv | AT josephbalderas acomparativestudyofdeepreinforcementlearningforcropproductionmanagement AT dongchen acomparativestudyofdeepreinforcementlearningforcropproductionmanagement AT yanbohuang acomparativestudyofdeepreinforcementlearningforcropproductionmanagement AT liwang acomparativestudyofdeepreinforcementlearningforcropproductionmanagement AT rencangli acomparativestudyofdeepreinforcementlearningforcropproductionmanagement AT josephbalderas comparativestudyofdeepreinforcementlearningforcropproductionmanagement AT dongchen comparativestudyofdeepreinforcementlearningforcropproductionmanagement AT yanbohuang comparativestudyofdeepreinforcementlearningforcropproductionmanagement AT liwang comparativestudyofdeepreinforcementlearningforcropproductionmanagement AT rencangli comparativestudyofdeepreinforcementlearningforcropproductionmanagement |