CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Ajuste de la caracterización foliar en manzanos realizada por un sistema de aplicación de pesticidas basado en sensores de ultrasonidos.

Estudiant que ha llegit aquest projecte:

Tutor/a o Cotutor/a: GIL MOYA, EMILIO

Departament: DEAB

Títol: Ajuste de la caracterización foliar en manzanos realizada por un sistema de aplicación de pesticidas basado en sensores de ultrasonidos.

Data inici oferta: 30-01-2023      Data finalització oferta: 30-09-2023

Estudis d'assignació del projecte:

Lloc de realització:


Paraules clau:
Sensor us, vegetación, fitosanitarios

Descripció del contingut i pla d'activitats:
Utilización de sensores de us para caracterización de la vegetación en frutales.

Reducción del uso de fitosanitarios

Overview (resum en anglès): The European Commission has implemented a set of policies with the purpose of promoting more sustainable and environmentally responsible agriculture, addressing the negative effects stemming from the use of phytosanitary products in agricultural activities. Through the strategies and guidelines of the European Green Deal, such as the 'Farm to Fork' strategy, the European Union has set the goal of reducing net greenhouse gas emissions by at least 55% by 2030, compared to the levels recorded in 1990.

Consequently, a detailed characterization of vegetation stands as an essential tool to drive more sustainable and efficient agriculture, aligned with the objectives and policies established in the European Green Deal. Research focused on the application of phytosanitary products is geared towards achieving more precise administration of these products through innovative designs and technologies that enable the application of appropriate doses in specific areas of crops, adjusting to the individual need of each plant. Among the employed technologies, LiDAR sensors, ultrasound systems, and depth cameras stand out.

However, until now, research conducted for vegetation characterization has faced difficulties in accurately comparing different types of sensors. This is due to the lack of evaluation of the accuracy and quality of captured data under same conditions and time. This study was focus on the evaluation and characterization of a multisensor platform composed by LiDAR sensor and an RGBD camera integrated into the ROS (Robot Operating System) for the characterization and assessment of density in an artificial apple orchard.

Through the comparative analysis between measurements obtained with the sensors and manually collected data, the accuracy of each sensor for tasks such vegetation characterization was evaluated. The results indicate that the LiDAR sensor presents higher precision in parameters such as tree height, vegetation height, and depth, with percentage of error ranging between 1% up to 2%. In contrast, the camera exhibits a margin of error that varies from 2% up to 8%. Both sensors showed a low magnitude of error, supporting their suitability for vegetation characterization and the consequent estimation of the volume needed to apply phytosanitary products.

LiDAR provides measurements with lower variability for vegetation height, tree height, width, and depth, in contrast to the camera, which is more consistent in porosity estimations.

The choice between the LiDAR sensor and the camera will depend on economic and infrastructure considerations, as well as the specific needs of the crop. Both sensors offer significant advantages in terms of precision, reliability and applicability, and the decision should be based on economic viability, which is conditioned by factors such as the type of crop, the size of the operation, and the associated equipment costs.

© CBLTIC Campus del Baix Llobregat - UPC