RObotics for MIcrofarms

Informazioni relative al progetto

ROMI

ID dell’accordo di sovvenzione: 773875

DOI 10.3030/773875

Progetto chiuso

Data della firma CE 3 Ottobre 2017

Data di avvio 1 Novembre 2017

Data di completamento 31 Luglio 2022

Finanziato da

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Costo totale € 3 868 186,25

Contributo UE € 3 868 186,25

Coordinato da

INSTITUT D'ARQUITECTURA AVANCADA DE CATALUNYA
Spain

Risultati finali

Documents, reports (16)

Intermediate dissemination report I

A description of the dissemination activities during the first period of the project.

Intermediate report on the theoretical advances and the experimental results of the adaptive learning system

Description of the theoretical advances during the covered period, and of the preliminary results of the experiments.

Intermediate report on the field studies II

The intermediate reports discuss the status of the field studies, describe the collected data sets and suggest amendments to the evaluation protocols.

Intermediate dissemination report II

A description of the dissemination activities during the second period of the project.

the DRONE + CABLE BOT (M34, M42, M48, M54)

1st version provided at M34 - update version at M42, M48, M54

Description of the experimental set-up of the outdoor field studies

This report describes the experimental set-up and the objectives of the outdoor field experiments for the LettuceThink robot and the NERO drone

Intermediate report on the field studies I

The intermediate reports discuss the status of the field studies, describe the collected data sets and suggest amendments to the evaluation protocols.

Simplified model of Arabidopsis thaliana

A model in OpenAleaLab demonstrating Arabidopsis development and its key control parameters.

Final dissemination report

A description of the dissemination activities during the last period of the project

the ROVER (M34, M42, M48, M54)

1st version provided at M34 - update version at M42, M48, M54

Report on the computer vision algorithms and their application

A description of the algorithms developed during the project and an evaluation of the results obtained

Detailed documentation of LettuceThink

The detailed description of LettuceThink user manual and building instructions

Final report on the field studies

The final report will provide the results of the evaluation of the weeding and crop monitoring application and provide the details of the data sets that were collected in the fields

Detailed documentation of NERO

The detailed description of NERO user manual and building instructions

the SCANNER 3D for scientific users (M34, M42, M48, M54)

1st version provided at M34 - update version at M42, M48, M54

Final report on the theoretical advances and the experimental results of the adaptive learning system

Full report on the theoretical developments during the project and detailed analysis of the results of the experiment

Demonstrators, pilots, prototypes (15)

Results of trained models

Software demonstrating the ability of the system to infer model parameter from observed data.

Second LettuceThink prototype

Updated prototype of the robot (hardware, software, weeding app).

Realistic model of Arabidopsis based on botanical measurements and extension to new plant species

Adaptive learning for the LettuceThink robot

Demonstration of the learning algorithms on the LettuceThink robot in an indoor environment

NERO aerial robotic solution (drone + cablebot) for microfarms

Final marketready version of the aerial robotic solution drone cablebot hardware software photogrammetry 3d scanning

Demonstration of enhanced 3d image segmentation for weeding

A weeding application integrating weeds detection based on 3d imaging.

Demonstration of 3d reconstruction of plants over time (4d) to follow plant growth

A crop monitoring app capable of following the 3d structure of selected plants over time

Development and assessment of a validation method for assessment of plant trait extraction from the ROMI pipeline

First NERO prototype

Demonstrator of the drone with selected cameras and modular frame size.

Second NERO prototype

aerial robotic solution (drone + cablebot) with 3d scanning and VR visualisation.

Demonstration of a 3d image segmentation application for weeding

A weeding application which depends solely on segmentation and path planning (starting from a clean bed).

LettuceThink robot

Final, market-ready version of the robot (hardware, software, weeding app, documentation).

First LettuceThink prototype

Early demonstrator of the robot (hardware, software, weeding app).

"Model of crop species #1"

A model in OpenAleaLab demonstrating a first crop species development and its key control parameters.

Demonstration of indoor and outdoor, 3d reconstruction of plants

A crop monitoring app capable of capturing the 3d structure of selected plants.

Open Research Data Pilot (1)

Data Management Plan

A detailed description of how to manage data during and after the project.

Pubblicazioni

Conference proceedings (7)

Transferring PointNet++ Segmentation from Virtual to Real Plants

Autori: Ayan Chaudhury, Peter Hanappe, Romain Azaïs, Christophe Godin, David Colliaux
Pubblicato in: ICCV 2021 - International Conference on Computer Vision, 2021
Editore: ICCV 2021

Ego-Noise Predictions for Echolocation in Wheeled Robots

Autori: Antonio Pico Villalpando, Guido Schillaci, Verena V. Hafner, Bruno Lara Guzmán
Pubblicato in: The 2019 Conference on Artificial Life, 2019, Pagina/e 567-573, ISBN 978-0-262-35844-6
Editore: MIT Press
DOI: 10.1162/isal_a_00222

3D Plant Phenotyping: All You Need is Labelled Point Cloud Data

Autori: A. Chaudhury, F. Boudon and C. Godin
Pubblicato in: Proceedings of Computer Vision Problems in Plant Phenotyping, 2020
Editore: CVPPP

Automated extraction of phyllotactic traits from Arabidopsis thaliana.

Autori: Wintz T., Colliaux D., Hanappe P
Pubblicato in: Workshop on Computer Vision Problems in Plant Phenotyping (CVPPP),, 2018
Editore: plant-phenotyping.org

A Deep Convolutional Neural Network Model for Sense of Agency and Object Permanence in Robots

Autori: Claus Lang, Guido Schillaci, Verena V. Hafner
Pubblicato in: 2018 Joint IEEE 8th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob), 2018, Pagina/e 257-262, ISBN 978-1-5386-6110-9
Editore: IEEE
DOI: 10.1109/devlrn.2018.8761015

Predictive Models for Robot Ego-Noise Learning and Imitation

Autori: Antonio Pico Villalpando, Guido Schillaci, Verena V. Hafner
Pubblicato in: 2018 Joint IEEE 8th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob), 2018, Pagina/e 263-268, ISBN 978-1-5386-6110-9
Editore: IEEE
DOI: 10.1109/devlrn.2018.8761017

Computational Agroecology: Should we bet the microfarm on it?

Autori: COLLIAUX, D., MINCHIN, J., GOELZER, S., AND HANAPPE, P
Pubblicato in: Eighth Workshop on Computing within Limits 2022, 2022
Editore: LIMITS

Other (6)

Creating smart and flexible farming tools for complex and biologically diverse environments

Autori: Cordis
Pubblicato in: Cordis, Numero 17 November 2020, 2020
Editore: Cordis

Developing Technology Towards Sustainable Farming Practices

Autori: Minchin. J
Pubblicato in: Open Access Government, Numero January Edition, 2020, Pagina/e 364 - 365, ISSN 2516-3817
Editore: Open Access Governmnet

House of Beautiful business : The digital and ecological transition

Autori: Minchin. J
Pubblicato in: House of Beautiful business : The digital and ecological transition, Numero 18 October 2020, 2020
Editore: House of Beautiful Business

Casa De Mateus : The digital and ecological transition

Autori: Minchin. J
Pubblicato in: Casa De Mateus : The digital and ecological transition, Numero January 15th 2020, 2020
Editore: Casa De Mateus

Automated vegetable growth analysis from outdoor images acquired with a cablebot.

Autori: Sollazzo A., Colliaux D., Garivani S., Minchin J., Garlanda L. and Hanappe, P.
Pubblicato in: www.plant-phenotyping.org, 2020
Editore: https://www.plant-phenotyping.org/

"""Horizon 2020/ Horizon Europe in the fields of Agriculture/ Rural & Digital – Lessons learnt, regulatory challenges, and looking ahead"""

Autori: Minchin. J
Pubblicato in: """Horizon 2020/ Horizon Europe in the fields of Agriculture/ Rural & Digital – Lessons learnt, regulatory challenges, and looking ahead""", Numero January 6 2020, 2020
Editore: European Commission

Peer reviewed articles (4)

The Weight Function in the Subtree Kernel is Decisive

Autori: Azaïs, Romain; Ingels, Florian
Pubblicato in: https://hal.archives-ouvertes.fr/hal-02097593, Numero 1, 2020, ISSN 1532-4435
Editore: MIT Press

Prerequisites for an Artificial Self

Autori: Verena V. Hafner, Pontus Loviken, Antonio Pico Villalpando, Guido Schillaci
Pubblicato in: Frontiers in Neurorobotics, Numero 14, 2020, ISSN 1662-5218
Editore: Frontiers Research Foundation
DOI: 10.3389/fnbot.2020.00005

Skeletonization of Plant Point Cloud Data Using Stochastic Optimization Framework

Autori: Ayan Chaudhury, Christophe Godin
Pubblicato in: Frontiers in Plant Science, Numero 11, 2020, ISSN 1664-462X
Editore: Frontiers Media S. A.
DOI: 10.3389/fpls.2020.00773

Intrinsic motivation and episodic memories for robot exploration of high-dimensional sensory spaces

Autori: Guido Schillaci, Antonio Pico Villalpando, Verena V Hafner, Peter Hanappe, David Colliaux, Timothée Wintz
Pubblicato in: Adaptive Behavior, 2020, Pagina/e 105971232092291, ISSN 1059-7123
Editore: SAGE Publications
DOI: 10.1177/1059712320922916

Book chapters (2)

Approximation of trees by self-nested trees

Autori: Romain Azaïs, Jean-Baptiste Durand, Christophe Godin
Pubblicato in: 2019 Proceedings of the Twenty-First Workshop on Algorithm Engineering and Experiments (ALENEX), 2019, Pagina/e 39-53, ISBN 978-1-61197-549-9
Editore: Society for Industrial and Applied Mathematics
DOI: 10.1137/1.9781611975499.4

Geometry Reconstruction of Plants

Autori: A. Chaudhury and C. Godin
Pubblicato in: 2020
Editore: CRC Press/Taylor and Francis