Skip to main content
ARS Home » Southeast Area » Mississippi State, Mississippi » Crop Science Research Laboratory » Genetics and Sustainable Agriculture Research » Research » Publications at this Location » Publication #424571

Research Project: Dynamic, Data-Driven, Sustainable, and Resilient Crop Production Systems for the U.S.

Location: Genetics and Sustainable Agriculture Research

Title: Performance simulation of real-time vision-based variable rate precision spray

Author
item Yao, Haibo
item TAIN, LEI - University Of Illinois Chicago
item TANG, LIE - Iowa State University
item STEWARD, BRIAN - Iowa State University
item Huang, Yanbo

Submitted to: Agricultural Engineering International: CIGR Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/31/2025
Publication Date: 3/30/2026
Citation: Yao, H., Tain, L., Tang, L., Steward, B., Huang, Y. 2026. Performance simulation of real-time vision-based variable rate precision spray. Agricultural Engineering International: CIGR Journal. 28(1):104-111. https://cigrjournal.org/index.php/Ejounral/article/view/10021/

Interpretive Summary: Herbicide is widely used for weed control in crop production. Herbicide can also bring in significant negative impacts and waste if over applied. Variable rate herbicide application is increasing being adopted in the industry with the advancement of precision agriculture technology. In a simulation study, this paper looks at the total amount of potential herbicide savings with variable rate applications. The results could provide suggestions in determining variable rate application strategies for precision weed management in crop production.

Technical Abstract: One of the main methods for weed control in crop field is through herbicide application. With technology development in sensing and control, variable rate (VR) herbicide application is increasing being adopted in the industry. In postemergence VR herbicide applications, machine vision is commonly used for weed detection and identification. The recent development of deep learning and artificial intelligence has greatly improved the accuracy and efficiency for weed detection, which makes it possible for more cost-effective VR herbicide applications. In this paper, spraying simulation models were developed to simulate real-time machine vision-based variable rate precision spraying. The objective was to examine the influence of different design factors and spray methods on the performance of VR precision spraying, e.g., total amount of herbicide saving over the uniform application method. Different sprayer travel speed and different control zone size were used in the models. The simulated spray methods include on/off intermittent spray, variable rate spray, and uniform low-dosage base rate plus variable rate spray. The results show that travel speed has no influence on herbicide saving. For the two variable rate spray methods, herbicide saving decreases when control zone size increases. For the on/off intermittent spray method, there is no difference in herbicide saving with different control zone sizes. Overall, for the on/off intermittent spray method, around 10% of herbicide could be saved while the variable rate spray method could save up to 45% of herbicide over the uniform spray method.