[Defended thesis] Anice Cheraiet: Experimental and statistical modeling of relations between morphological characteristics of grapevine and spraying deposits: application to precision agriculture

Anice Cheraiet is one of the #DigitAg co-funded PhDs

He defended his thesis on Friday 18th December.

Experimental and statistical modeling of relations between morphological characteristics of grapevine and spraying deposits: application to precision agriculture

  • Start Date: January 2018
  • University: MUSE Montpellier University of Excellence / Institut Agro – Montpellier SupAgro
  • PhD School: GAIA , MUSE Montpellier Université d’Excellence
  • Field(s): Functional ecology – Precision viticulture – Bio statistics – Biophysical measurements
  • Doctoral Thesis Advisor: Carole Sinfort (Montpellier SupAgro, UMR Itap)
  • Co-supervisors : Bernadette Ruelle (Inrae, Itap), Olivier Naud (Inrae, Itap), Sébastien Codis (IFV), Nadine Hilgert (Inrae, Mistea)
  • Funding: #DigitAg – IFV Acta-les instituts techniques
  • #DigitAg: Co-funded PhD – Challenge 3

Keywords: Precision spraying, Architecture of the plant, Agro-processes, Modelization, Reduction of phytosanitary inputs, Dose

Abstract: The societal demand for reducing pesticides use and the context resulting from European directive 2009/128/EC and French action plan Ecophyto bring forward the need to reconsider plant protection processes. One issue raised is the adaptation of sprayer set points (technological parameters, dose) with regard to morphological characteristics of grapevine and according to sprayer. The performance of a spraying process can be described by its efficiency, namely the quantity of product that reaches target for a given sprayed quantity, and by its quality, which is essentially the homogeneity of deposits by surface unit within the canopy. The objective of the thesis is to develop models that predict quantities and profiles of deposits on vegetation with regard to canopy morphology and characteristics of sprayers, within a precision agriculture framework. LiDAR technologies, in particular, will be mobilized in order to characterize this morphology and propose indicators. The modeling framework will be multi-scale. The applied objective is to analyze the interest of developing innovative precision spraying techniques associated to a numerical and data intensive agriculture. Different technological scenarios will be considered and the potential savings in chemical inputs, with regards to classical spraying at homologated dose, will be evaluated. The methodological questions involved in experimental and statistical modeling will be as follows: (I) combination of high resolution (LiDAR) and ponctual (readings) measures, as well as physical sampling (deposits), (ii) adaptation of observation and decision to different technological scenarios and different scales, (iii) sensibility analysis for accuracy and variability of indicators with regard to sampling configuration and observed structures of spatial correlation. Inversion of models should make it possible to compute recommendation maps for sprayer set points from mapsdescribing the canopy.

Papers at international conferences

Download the thesis report

A. Cheraiet, M. Carra, A. Lienard, S. Codis, A. Vergès, X. Delpuech, O. Naud (2019), Investigation on LiDAR based indicators for predicting agrochemical deposition within a vine field, Precision Agriculture 2019, Proceedings of the 12th European Conference on Precision Agriculture – https://dx.doi.org/10.3920/978-90-8686-888-9_18

A. Cheraiet, X. Delpuech, M. Carra, J. Andres, A. Vergès, A. Lienard, S. Codis, O. Naud. (2019), Evaluate sensors and innovative digital solutions in the vineyard to reduce and manage phytosanitary inputs, Environmental Science And Pollution Research

Cheraiet A., Naud O., Carra M., Codis S., Lebeau F., Taylor J. (2020) An algorithm to automate the filtering and classifying of 2D LiDAR data for site-specific estimations of canopy height and width in vineyards, Biosystems Engineering

Contact:  anice.cheraiet [AT] vignevin.com​ – Tél : 0679929173

Réseaux : ResearchGate