Description du projet
Des ordinateurs pour quantifier les récoltes
L’estimation du rendement des cultures est une fonction centrale de la production fruitière car elle détermine la part de la production qui convient à la consommation et à la commercialisation. L’estimation du rendement, qui est cruciale par exemple pour la production de pommes, doit être fiable, rapide et rentable. Aujourd’hui, les fruits sont comptés manuellement par les ouvriers agricoles. Non seulement cela prend beaucoup de temps, mais la précision ne peut pas être garantie, surtout pour les gros rendements. Le projet AGERPIX, financé par l’UE, travaille sur une solution qui repose sur des algorithmes d’intelligence artificielle. Le système devrait permettre de disposer de prévisions fiables et rapides tout en réduisant considérablement les coûts d’exploitation et de gestion. Il fournira aux producteurs des informations sur la taille et la qualité des fruits et sur le rendement. Il calculera en outre les besoins en main d’œuvre et proposera une planification logistique.
Objectif
"Crop yield estimation is an important task in apple orchard management. The current practice of yield estimation is based on manual counting of fruits by workers. It is extremely time-consuming, labour-intensive, highly inaccurate, and it is not practical for large fields. Agerpix provides accurate predictions to help growers improve fruit quality and reduce operating costs by making better decisions on intensity of fruit thinning and plant nutrients and treatments (mid-season), size of the harvest labour force, machinery and materials and logistical planning of storage, packing and cold warehouses, not to mention the development of a commercialization strategy tailored to the expected production, achieving a 50% cost reduction in orchard management operations. Artificial Intelligence algorithms are used to identify, measure diameter ranges, and envision the fruit leafiness and vigour, providing yield estimations over the plant heights and plant health variables.
Several piloting projects for the yield estimation system at top apple producers (Nurfri - #1 Spanish and Blue Whale #1 French among others) have been deployed with +140ha analysed with a precision of 90-95%. AGERPIX system has been adapted to four different apple varieties. After successful validation activities, CODESIAN is developing a customer portfolio worth 38M€ in five years. However, because AGERPIX is offering as a B2B service and because the technology can be easily replicated to other fruits (ongoing validations with table grapes and peach with minor AI/sensor adaptations), a careful scale-up design to strengthen the business plan towards covering global needs fruit markets (apple: 517 M€; table grape: 124 M€; peach: 160M€; tangerine: 299 M€; avocado: 54 M€) is needed. After further data gathering through extensive validations across new fruits, CODESIAN projects +7,9M€ revenues with +4,5M€ EBIT and +60 new jobs created by 2024.
"
Champ scientifique
Not validated
Not validated
- natural sciencescomputer and information sciencesartificial intelligence
- agricultural sciencesagriculture, forestry, and fisheriesagricultureagronomyplant protection
- natural sciencesbiological scienceszoologymammalogycetology
- agricultural sciencesagriculture, forestry, and fisheriesagriculturehorticulturefruit growing
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensors
Programme(s)
Régime de financement
SME-1 - SME instrument phase 1Coordinateur
42190 GOLMAYO (SORIA)
Espagne
L’entreprise s’est définie comme une PME (petite et moyenne entreprise) au moment de la signature de la convention de subvention.