Skip to main content
ARS Home » Plains Area » Bushland, Texas » Conservation and Production Research Laboratory » Soil and Water Management Research » Research » Publications at this Location » Publication #405169
Research Project: Dryland and Irrigated Crop Management Under Limited Water Availability and Drought

Location: Soil and Water Management Research

Title: Deep reinforcement learning-based irrigation scheduling

Author
item YANG, YANXIANG - Amazon Web Services
item HU, JIANG - Texas A&M University
item PORTER, DANA - Texas A&M University
item MAREK, THOMAS - Texas A&M Agrilife
item HEFLIN, KEVIN - Texas A&M Agrilife
item KONG, HONGXIN - Texas A&M University
item SUN, LIJIA - Texas A&M University

Submitted to: American Society of Agricultural and Biological Engineers
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/15/2019
Publication Date: 3/5/2020
Citation: Yang, Y., Hu, J., Porter, D., Marek, T., Heflin, K., Kong, H., Sun, L. 2020. Deep reinforcement learning-based irrigation scheduling. American Society of Agricultural and Biological Engineers. 63(3):549-556. https://doi.org/10.13031/trans.13633.
DOI: https://doi.org/10.13031/trans.13633

Interpretive Summary: A computational process called machine learning has been widely applied in many areas, with promising results. However, machine learning has not been applied widely to irrigation scheduling. Compared with traditional reinforcement machine learning, a deep reinforcement learning method can better model a real-world environment. Working in a project funded by the USDA ARS Ogallala Aquifer Program, scientists from Texas A&M AgriLife proposed a deep reinforcement learning-based irrigation scheduling process. Simulations for various weather conditions and crop types showed that the proposed deep reinforcement learning irrigation scheduling might increase net return. These results are of interest to irrigation equipment manufacturers and irrigators.

Technical Abstract: Machine learning has been widely applied in many areas, with promising results and large potential. In this article, deep reinforcement learning-based irrigation scheduling is proposed. This approach can automate the irrigation process and can achieve highly precise water application that results in higher simulated net return. Using this approach, the irrigation controller can automatically determine the optimal or near-optimal water application amount. Traditional reinforcement learning can be superior to traditional periodic and threshold-based irrigation scheduling. However, traditional reinforcement learning fails to accurately represent a real-world irrigation environment due to its limited state space. Compared with traditional reinforcement learning, the deep reinforcement learning method can better model a real-world environment based on multi-dimensional observations. Simulations for various weather conditions and crop types show that the proposed deep reinforcement learning irrigation scheduling can increase net return.