Project Number: 6064-21660-001-033-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Oct 1, 2018
End Date: Sep 30, 2022
ARS Objective: Develop advanced UAS/UAV application systems and data management systems and Bioinformatics tools that integrate developed G x E x M data into precision agricultural crop management for Mid-South crops. The systems and tools should lead to improvements in agricultural productivity and agricultural system landscape management. Objectives and approaches to be developed jointly between ARS and MSU Objectives: To cooperate in the development of advanced UAS/UAV application systems and data management systems and bioinformatics tools that integrate GxExM data into precision agricultural crop management for regional relevant crops. The systems and tools should lead to improvements in agricultural productivity and agricultural landscape management. Geosystems Research Institute of Mississippi State University will develop advanced UAS\UAV applications and data management systems and Bioinformatics tools that integrate Genetics x Environment x Management (G x E x M) data into precision agricultural crop management for Mid-South crops. Geosystems Research Institute will cooperate with ARS by: A) Develop the best practices for collecting imagery and data with UAS payloads to detect phenomenon within field that indicate plant stress and/or soil health as relates to research undertaken. B) Develop algorithms that automate UAV image processing, for use with in-situ data as appropriate, to detect phenomenon within field that indicate plant stress and/or soil health. Develop algorithms to detect and quantify different types of plant stress and/or soil health using multiple sensors. C) Develop decision support tools that incorporate imagery collected according to best practices with developed algorithms that can be used to optimize sustainability within Mid-south agricultural production systems.
Gather UAV data at multiple altitudes, with multiple sensors, and at multiple times within the crop growth life cycle for the purposes of developing the best practices for detection of phenomenon within field that is indicative of plant stress and/or soil health, as relates to research undertaken by USDA under Objectives 1 and 2 of ARS parent project. Imagery will be evaluated based on the potential to provide reliable data that form the basis for actionable information, required by end users. Understandably, these will not be one-size fits all solutions and may require deviation from status quo UAS operating procedures. Images will be analyzed using the most appropriate techniques for the purposes of creating automated routines that utilize supervised and unsupervised classification techniques such as deep learning, and support vector machines. Based on the outcomes of image analysis, algorithms will be created to rapidly process data from future missions. At the same time, a systems modeling effort will be undertaken in support of decision support tool creation. With successful completion of these three tasks, end users should be capable of collecting the necessary input data, with the UAV and in-field observation, required for automated detection of phenomenon within field that is indicative of plant stress and/or soil health. When coupled with decision support tools, predictive and prescriptive actions to optimize sustainability within Mid-south production systems will be viable.