LiDAR-based perception for agricultural autonomous robots

• Grant number: 18/10894-2
• Support Opportunities: Scholarships in Brazil - Doctorate (Direct)
• Start date: March 01, 2019
• End date: February 28, 2021
• Field of knowledge: Engineering - Mechanical Engineering
• Principal Investigator: Marcelo Becker
• Grantee: Vitor Akihiro Hisano Higuti
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated research grant: 13/07276-1 - CEPOF - Optics and Photonic Research Center, AP.CEPID

Abstract

Lightweight ground robots have been increasingly showing their potential to tackle agricultural tasks that require time-consuming human labor thus limited in detail or area coverage. For example, autonomous robots able to safely traverse the cropland could carry numerous devices to collect information about soil or even from each plant. However, a reliable autonomous navigation is still a challenge. A robot needs to deal with the agricultural field dynamism besides unpredictable obstacles, e.g. humans, animals and machines. In particular, corn and soybean fields present additional issues due to the standard way of cultivation: narrow lanes (less than a meter) that become even less visible by later stages because of dense coverage and spreading of leaves. These conditions heavily influence the sensors by provoking frequent occlusions, misreadings and by putting them under situations out of their workingrange. Therefore, this project proposes the development of a perception subsystem that filters, classifies and mathematically converts raw sensor readings into a compact knowledge stream for reliable autonomous navigation. To do that, the core sensingdevice is a Hokuyo UTM30-LX LiDAR, whose 2-D scans will be placed in 3-D space with the assistance of IMU and odometry data. By using the proposed perception subsystem as the input of autonomous navigation methods, it will be evaluated with extensive field experiments in corn and soybean crops. (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)

ARAUJO, GABRIEL L.; JORGE, ID E. FILHO; HIGUTI, VITOR A. H.; BECKER, MARCELO; DASILVA, BMF; TODT, E; NASCIMENTO, TP; DOSSANTOS, DH; CURVELO, CDF; FABRO, JA. A New Approach of Monocular Visual Odometry to Trajectory Estimation Within a Plantation. 2021 LATIN AMERICAN ROBOTICS SYMPOSIUM / 2021 BRAZILIAN SYMPOSIUM ON ROBOTICS / 2021 WORKSHOP OF ROBOTICS IN EDUCATION (LARS-SBR-WRE 2021), v. N/A, p. 6-pg., 2021-01-01. (20/11089-6, 20/10533-0, 18/10894-2, 13/07276-1)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

HIGUTI, Vitor Akihiro Hisano. 2D LiDAR-based perception for under canopy autonomous scouting of small ground robots within narrow lanes of agricultural fields. 2021. Doctoral Thesis - Universidade de São Paulo (USP). Escola de Engenharia de São Carlos (EESC/SBD) São Carlos.