Intelligent systems: predictive modeling and internet of things for animal production in agriculture 4.0

• Grant number: 22/07442-8
• Support Opportunities: Regular Research Grants
• Start date: November 01, 2022
• End date: July 31, 2025
• Field of knowledge: Agronomical Sciences - Agricultural Engineering - Rural Buildings and Ambience
• Agreement: CONFAP - National Council of State Research Support Foundations
• Principal Investigator: Iran José Oliveira da Silva
• Grantee: Iran José Oliveira da Silva
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Associated researchers: Glauber da Rocha Balthazar ; Maira Martins da Silva ; Sergio Luis de Castro Junior ; Thiago Kraetzer ; Thiago Luís Lopes Siqueira
• Associated scholarship(s): 24/15425-1 - Data Science to connect digital farming and eScience: from smart farms to research infrastructures, BP.IC

Abstract

This project is the result of cooperation between researchers in the areas of animal environment. Precision animal farming and animal welfare are associated with the development of products, services, and innovation, enabled by technologies aimed at digital agriculture within the agricultural value chain, emphasizing software and hardware that provide solutions and adjustments in the stages of agricultural production processes. It encompasses precision livestock, farm automation and robotics, computational methods, image processing, computer vision, intelligent systems, big data techniques, and the internet of things. Researchers from the University of São Paulo - Luiz de Queiroz Higher School of Agriculture (ESALQ/USP) and the Federal Rural University of Pernambuco (UFRPE) will work together. The purpose of the proposal is to apply digital technologies and deep machine learning techniques to automate complex handling processes in animal production, increasing their productivity and efficiency through a variety of disruptive technologies that reach agriculture and can help producers manage their production units more safely and sustainably. Likewise, proposed management of the production system using the principles of Internet of Things (IoT), robotics and machine learning, and Artificial Intelligence (AI). The research will be divided into four stages. The first two will be carried out at UFRPE, namely: 1. Study and develop automatic predictive techniques to assist in the early diagnosis of bovine mastitis from sets of digital thermal images of animals with and without a known diagnosis; 2. Develop an intelligent IoT system capable of automatically monitoring cattle behavior on pasture. The subsequent steps will involve the poultry chain developed by the ESALQ/USP Team. Using IoT, robotics, and Machine Learning (ML) tools to develop, validate and apply an animal-robot for hatchery and aviary environments (specifically robot-egg and robot-chicken), characterizing it as a biosensor agent in the acquisition of environmental variables in a continuous way, allowing to collect, transfer and analyze the information in real-time, with the presentation of predictive algorithms. The specific objectives are 3. Develop a monitoring system to acquire data from the internal environment of incubators and inside fertile eggs associated with machine learning for decision-making and reduction of productive losses in hatcheries (robot-egg). 4. Develop a robotic and mobile model in the production environment (broiler aviary) recording environmental information in real-time using sensory devices for the acquisition and transfer of data for use in machine learning and generation of predictive information for problem-solving in industrial poultry (robot-chicken). It is expected that these research actions can help in the transfer of technologies applied to artificial intelligence in animal production and information relevant to the improvement of local productive arrangements in the states of Pernambuco and São Paulo and consequently for the whole country. (AU)

Scientific publications

(5)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

GUESINE, GIOVANE DEBS; SILVEIRA, ROBSON MATEUS FREITAS; DA SILVA, IRAN JOSE OLIVEIRA. Physical and environmental characteristics of the compost barn system and its effects on the physical integrity, reproduction and milk production of dairy cattle: a scoping review. JOURNAL OF ANIMAL BEHAVIOUR AND BIOMETEOROLOGY, v. 11, n. 2, p. 6-pg., 2023-01-01. (22/07442-8)

DE CASTRO JUNIOR, SERGIO LUIS; BALTHAZAR, GLAUBER DA ROCHA; SILVEIRA, ROBSON MATEUS FREITAS; DA SILVA, IRAN JOSE OLIVEIRA. Multilevel sensor for monitoring external and internal environment of eggs. POULTRY SCIENCE, v. 103, n. 7, p. 12-pg., 2024-05-14. (22/07442-8)

VIEIRA, FERNANDA VICTOR RODRIGUES; SILVEIRA, ROBSON MATEUS FREITAS; FRANCHI, GUILHERME AMORIM; DA SILVA, IRAN JOSE OLIVEIRA. The impact of training on stockpersons' behaviour and cows' fear response. JOURNAL OF ANIMAL BEHAVIOUR AND BIOMETEOROLOGY, v. 11, n. 2, p. 8-pg., 2023-04-01. (22/07442-8)

BALTHAZAR, GLAUBER DA ROCHA; SILVEIRA, ROBSON MATEUS FREITAS; DA CRUZ, MARCOS VINICIUS AMATO; DA SILVA, IRAN JOSE OLIVEIRA. Development of an electromechanical device for real-time detection of litter moisture in commercial broiler. SMART AGRICULTURAL TECHNOLOGY, v. 12, p. 13-pg., 2025-12-01. (22/07442-8)

DA SILVA, ALLAN BASTOS; LOPES SIQUEIRA, THIAGO LUIS. Design, Implementation and Evaluation of an Information System for the Management of Digital Farming Scientific Data. 2024 L LATIN AMERICAN COMPUTER CONFERENCE, CLEI 2024, v. N/A, p. 10-pg., 2024-01-01. (22/07442-8)