Adaptive LPV ARMAX model in predictive flow control based on variable fluidic resistance for agricultural sprayers

• Grant number: 25/03762-6
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: July 28, 2025
• End date: July 27, 2026
• Field of knowledge: Engineering - Electrical Engineering - Industrial Electronics, Electronic Systems and Controls
• Principal Investigator: Vilma Alves de Oliveira
• Grantee: Deniver Reinke Schutz
• Supervisor: Yiannis Ampatzidis
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Southwest Florida Research & Education Center, United States
• Associated to the scholarship: 23/06759-0 - Adaptive predictive control for land sprayers in variable rate application, BP.DR

Abstract

This project focuses on the development and implementation of an advanced platform for flow and pressure control in agricultural sprayers. The objective of the project takes into account the incorporation of parallel coordination of nozzles and intelligent system controllers to optimize application quality, as well as reduce operating costs. In addition, a strong emphasis on safety, environmental sustainability and health is considered. In fact, preliminary results have demonstrated non-linear variations in the mathematical models, leading to a reduction in the efficiency of the spraying system, that is, motivating this research and development. As a doctoral research, this project is being developed at the University of São Paulo in partnership with Embrapa Instrumentation and a significant part of it will be completed at the Southwest Florida Research and Education Center - University of Florida. This opportunity will allow not only better access to advanced knowledge in agricultural pest control systems, but also research collaboration with one of the best agricultural groups in the United States. In addition, during the period at the University of Florida, a focus on the development of an ARMAX model with Linear Parameter Variation (LPV) characteristics will be prioritized and developed. Therefore, such a development will allow the predictive controller to adapt the system for better nozzle switching, particularly due to variations in nozzle fluid resistance values. Onboard testing will be carried out on a robotic smart sprayer system, enabling field experiments, spot spraying, as well as smart sprayers in a real agricultural field. This initiative will enable the delivery of the advanced platform for sprayers using predictive control techniques and computational intelligence to aid decision making, as well as will consider validation using not only data obtained in a simulation environment, but mainly experimental data obtained in real crop areas. (AU)