Monitoring of Sugarcane Biophysical Attributes Using UAVs: Integration of LiDAR and Multispectral Data

• Grant number: 25/14887-4
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: November 01, 2025
• End date: February 29, 2028
• Field of knowledge: Agronomical Sciences - Agronomy
• Principal Investigator: Ana Cláudia dos Santos Luciano
• Grantee: Matheus Sterzo Nilsson
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil

Abstract

Sugarcane is one of the main crops in Brazilian agribusiness, requiring technologies that optimize the monitoring of its development and biomass estimation. Currently, the manual collection of biophysical attributes-such as plant height, number of stalks per linear meter, leaf area index (LAI), and biomass-provides important information for breeders and producers. However, this process is time-consuming, costly, and has low scalability. Unmanned aerial vehicles (UAVs) equipped with sensors have increasingly been used as an alternative to overcome these limitations in manual field surveys. Nevertheless, further studies are needed to integrate data from different UAV-mounted sensors for the prediction of multiple biophysical attributes in commercial areas with high environmental variability. This research aims to integrate LiDAR and multispectral data from UAVs to estimate sugarcane biophysical attributes (plant height, stalk diameter, number of stalks per plant, and LAI) at different stages of crop development. The methodology will be divided into three main stages: (i) data acquisition in the field and with UAVs equipped with multispectral and LiDAR sensors, aiming to extract spectral and structural variables related to sugarcane; (ii) integration of field and sensor data followed by the application of machine learning algorithms to estimate biophysical attributes; and (iii) assimilation of these estimates into a mechanistic crop growth model (APSIM), in order to evaluate the impact of the estimated variables on crop development simulation and yield prediction. The expected results include the development of empirical models based on spectral, structural, and textural variables capable of accurately estimating sugarcane biophysical parameters. The study also seeks to assess the performance of different input data combinations in biomass estimation using empirical models and to verify whether UAV-derived data from different sensors can be assimilated into mechanistic models to improve crop growth simulation and provide more accurate yield forecasts under real production conditions. The advances achieved may contribute to improving monitoring and management practices in sugarcane production, directly benefiting operational efficiency and strategic planning in the sector. (AU)