Identification of entrances and exits on trails in medium-sized crop fields through computer vision

• Grant number: 20/13037-3
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2021
• End date: April 30, 2022
• Field of knowledge: Engineering - Mechanical Engineering
• Principal Investigator: Marcelo Becker
• Grantee: Estêvão Serafim Calera
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The world's population has grown at an exponential rate due to the increase in life expectancy as a result of technological advances in recent centuries. Food production needs to keep up with this growth and encounters barriers in increasingly scarce demographic spaces and inefficient human labor. In this context, autonomous robotics is fundamental, allowing the acquisition of data and the performance of tasks with greater precision and agility. A robust autonomous navigation system implies the development of several sections of the route, one of which being the starting and ending areas of the crop trails. These regions are often bushy, what can occlude the reading of sensors. Therefore, in this research, methods based on Computer Vision will be used to identify exits and entrances in the plantation corridors by analyzing the color spectra in different regions of the images obtained via TerraSentia robot cameras. Using this algorithm, decision-making is expected to be accurate and it would provide a reliable direction for robot navigation. (AU)

Scientific publications

(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)

CALERA, ESTEVAO SERAFIM; DE OLIVEIRA, GABRIEL CORREA; ARAUJO, GABRIEL LIMA; FACURI FILHO, JORGE ID; TOSCHI, LUCAS; HERNANDES, ANDRE CARMONA; BAQUERO VELASQUEZ, ANDRES EDUARDO; GASPARINO, MATEUS VALVERDE; CHOWDHARY, GIRISH; HISANO HIGUTI, VITOR AKIHIRO; et al. Under-Canopy Navigation for an Agricultural Rover Based on Image Data. JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS, v. 108, n. 2, p. 18-pg., 2023-06-01. (20/10533-0, 20/11262-0, 20/13037-3, 20/11089-6)