Go to contentGo to menuGo to searchGo to the news list

Laboratoire Angevin de Recherche en Ingénierie des Systèmes


Main navigation

    Search

    Breadcrumb

    POODLE

    POODLE

    • Share this page on social networks
    • E-mail this page

      Send by mail


      Separated by coma
    • Print this page

    Research project POODLE

    Prototype Laser Weeding

     

    Teams: 

    Dynamic and discrete events and Optimization

    and

    Information, Signal, Image Processing and Life Sciences

     

    Labeling: none


    Term: 3 years (2017 - 2020)

     

    Funding: Atlanstic RFI 2020

     

    LARIS staff involved:

    Dynamic and discrete events and Optimization

    Rémy Guyonneau

     

    Information, Signal, Image Processing and Life Sciences

    Étienne Belin

     

    and Franck Mercier

     

    Project partners: Matthieu Loumaigne (Moltech - Angers), Tristan Boureau (IRHS - Angers)

     

     

    Project description

    The control of weeds (weeds) is a major challenge for agriculture, especially for so-called field crops (cereals, oilseeds, ...) or market gardening where competition between plants of interest and weeds. Currently, the lack of manpower and the increasingly restrictive use of phytosanitary products require the establishment of new alternative solutions for the control and weeding of weeds. Today, the use of robots is more and more studied and developed to carry out these actions of weeding. We can notably note the Oz robot of the Toulouse company Naïo Technologies or the BoniBob weeding robot of the Bosh company. In both cases, robots use mechanical weeding tools. Although these tools have been proven, they show their limits when weeds are surrounded by cultures of interest (eg dense crops). A promising alternative is the use of lasers, as studied in [1, 2, 3].

    Our project therefore proposes to focus on this technique made possible by recent technological advances. Miniaturization and research into semiconductors has made it possible to offer laser diodes that are both powerful enough and in a wide range of wavelengths. The approach envisaged consists of depositing the energy of the light radiation in a focused way on the weed, and potentially destroying it by thermal effect. This laser / material interaction is based on the choice of laser wavelength(s), in connection with the absorption parameters of the main molecules (water and chlorophyll) of the plants.

    We propose here to make a prototype "proof of concept" using this approach of weeding. We identified four main (non-prioritized) tasks related to interconnected disciplines:

    • photonics: realization of a laser (power, wavelength, focusing optics) able to destroy weeds,
    • plant / laser interaction: studies of the effects of laser light pulses on the physiology of weeds,
    • computer vision: detection and localization of weeds in a given scene by image processing from RGB sensor,
    • robotics: motorization and control of the laser (mechatronic design of the actuators).

    The coupling and integration of these four tasks in a first prototype will allow the realization of a robotic weeding system.

    For the identification of an effective laser and its effect on the physiology of weeds, the knowledge and skills of the Moltech Laboratory of the University of Angers and more particularly the Atomic and Molecular Spectroscopy, Surface Structuring and Optics (SAMSON) team will be exploited [4], together with those of the IRHS laboratory of INRA [5]. The LARIS' Information, Signal, Image and Life Sciences (ISISV) team will focus on the detection of weeds using image processing algorithms [5]. Finally, the LARIS' Dynamic Systems and Optimization (SDO) team will focus on the realization and control of the pointing tool [6].

    The deliverable envisaged at the end of the project consists of two prototypes: one to demonstrate the feasibility of the control in position of a laser beam and the other to evaluate the destructive effect of a laser beam on a weed.

     

    Bibliography

    [1] S.K. Mathiassen, et al. "The effect of laser treatment as a weed control method". Biosystems engineering, vol. 95, no 4, p. 497-505, 2006.

    [2] S. Kaierle, et al. "Find and irradiate-lasers used for weed control". Laser Technik Journal, vol. 10, no 3, p. 44-47, 2013.

    [3] R. Shah, et al. "An approach to a laser weeding system for elimination of in-row weeds". In: Precision agriculture'15. Wageningen Academic Publishers, p. 19-47, 2015.

    [4] M. Loumaigne, et al. "Optical extinction and scattering cross sections of plasmonic nanoparticle dimers in aqueous suspension". Nanoscale. Vol. 8 n°12 p. 6555-6570, 2016.

    [5] E. Belin, T. Boureau et al. "Thermography versus chlorophyll fluorescence imaging for detection and quantification of apple scab". Computers and Electronics in Agriculture, 90, 159-163, 2013.

    [6] R. Guyonneau, et al. "Guaranteed Interval Analysis Localization for Mobile Robots". Journal on Advanced Robotics. Vol. 28 n°16 p. 1067-1077, 2014.