List of public deliverables, papers, documents, and datasets
This document gives a view of our long-term vision and how the results of the different work packages come together in concrete scenarios.
Link:
ROMI-Vision_and_Scenarios-1.pdf
D1.1 - IPR Strategic Plan Reporting period 1
ROMI’s IPR strategy is based on an open-software/open-hardware model. In this document we give a more in-depth description of this approach. We present a bibliographic study and look at some of the related business models.
Link:
ROMI-D1.1-IPR_Strategic_Plan.pdf
Additional documents:
Annexes
D1.2 - The Romi Rover - Project Management Reporting period 2
This document covers the status of the Romi Rover, including the usage scenario, the development progress, the specifications, the user studies, market studies, and a draft user manual.
Link:
ROMI-D1.2-Management-document_-_The-Rover.pdf
WP2 - Robot for weeding and detailed crop monitoring
D2.1 - First LettuceThink prototype Reporting period 1
WP3 - Drone for aerial crop monitoring
D3.1 - First NERO prototype Reporting period 1
WP4 - Adaptive learning for the coordinated control of sensors and actuators
The goal of the work in WP4 system is to improve the 3D reconstruction of the plants by generating robot movements that optimise the information obtained by the camera sensors.
D4.1 - Report on the experimental set-up of the adaptive learning system Reporting period 1
D4.2 Intermediate report on the theoretical advances and the experimental results of the adaptive learning system Reporting period 2
WP5 - Advanced sensing and analysis of crops
D5.1 - Demonstration of a 3d image segmentation application for weeding Reporting period 1
D5.3 - Demonstration of indoor and outdoor, 3d reconstruction of plants
The ROMI project develops a toolbox for plant scanning and analysis that combines different approaches. In WP6 we develop an approach that is heavily based on plant models. These models are used as a support for the data analysis and plant segmentation but also as a source for synthetic images used to train neural networks.
D6.1 - Simplified model of Arabidopsis thaliana Reporting period 1
D6.2 - Model of crop species #1 Reporting period 2
During the second reporting period, we developed a first model of a crop species, notably the tomato plant. An improved version of the
A. thaliana model, with improved rendering, is also discussed.
Link:
ROMI-D6.2-Model-of-crop-species-1.pdf
Videos:
D6.3 - Results of trained models Reporting period 2
This document the results we obtained by using the model of
A.thaliana developed above to train neural networks to segment images of real
A. thaliana plants.
Link:
ROMI-D6.3-Results-of-trained-models.pdf
WP7 - Data acquisition and real-world application testing
D7.1 - Description of the experimental set-up of the outdoor field studies Reporting period 1
D7.2 - Intermediate report on the field studies I Reporting period 1
WP8 - Dissemination and exploitation
D8.1 - Data Management Plan Reporting period 1
D8.2 - Intermediate dissemination report I Reporting period 1
D9.1 - POPD - Requirement No. 1 Reporting period 1
Point clouds of Can Valldaura