Skip to main content
ARS Home » Plains Area » El Reno, Oklahoma » Oklahoma and Central Plains Agricultural Research Center » Agroclimate and Hydraulics Research Unit » Research » Publications at this Location » Publication #433828
Research Project: Development of a Monitoring Network, Engineering Tools, and Guidelines for the Design, Analysis, and Rehabilitation of Embankment Dams, Hydraulic Structures, and Channels

Location: Agroclimate and Hydraulics Research Unit

Title: AGBOT: Intelligent crop diagnosis and Large Language Model (LLM)-based decision support tool

Author
item BOYAPALLY, DHANYA - University Of Missouri
item KARRA, KRISHNA - University Of Missouri
item FRATER, JACK - University Of Missouri
item Hunt, Sherry
item ALOYSIUS, NOEL - University Of Missouri

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 4/10/2026
Publication Date: 4/20/2026
Citation: Boyapally, D., Karra, K., Frater, J., Hunt, S., Aloysius, N. 2026. AGBOT: Intelligent crop diagnosis and Large Language Model (LLM)-based decision support tool. Meeting Abstract. University of Missouri, Show Me Research Week, Columbia, MO, April 20-24,2026.

Interpretive Summary:

Technical Abstract: AGBOT is an AI-driven crop intelligence platform designed to support accurate plant disease and pest identification through an integrated computer vision and decision support system. The core of the platform is an EfficientNet-based deep learning model, which operates within a structured computer vision pipeline involving image preprocessing, normalization, and feature extraction. This enables the system to learn fine-grained visual patterns such as discoloration, lesions, and insect damage, allowing reliable classification across diverse plant conditions. To improve robustness in real-world usage, the system incorporates a fallback diagnostic mechanism based on a questionnaire-driven approach. When image input is unavailable or model confidence is low, the system collects symptom-based inputs from users and refines predictions using structured reasoning. This hybrid approach ensures consistent performance even under uncertain or incomplete data conditions. A Large Language Model (LLM) is integrated as a reasoning and interaction layer, transforming model outputs into clear, context-aware explanations and actionable treatment recommendations based on Integrated Pest Management principles. The LLM also enables interactive communication, allowing users to ask follow-up questions and receive adaptive guidance. Additionally, the platform supports multilingual and bilingual interaction, making it accessible to a wider range of users. It also allows customization through user-defined parameters such as farm size and crop type, enabling more relevant and context-aware recommendations. By combining efficient visual perception, adaptive reasoning, and user-centric design, AGBOT provides a scalable and intelligent solution for modern agricultural decision support.