Close
Currently the apple orchard has the Hesperides platform as digital field notebook (https://hesperides.farm/) for data logging, management and fungus prediction. It also has a meteorological station to measure rain, temperature, humidity and wind speed. For each apple variety there are humidity probes in the soil (at three levels), in order to measure the humidity and adjust the irrigation accordingly, once per day. Also, the orchard field has poles every 10.8 meters that are georeferenced with centimeter precision.
In terms of spraying, it is done with human-driven tractors and they are equipped with a GNSS system to provide localization information only.
Farmers’ and their associations
The apple orchard has an extension of 130 hectares and produces: Royal Gala, Golden Delicious, Red Delicious, Granny Smith, Fuji and Pink Lady.
In the context of the Smart Droplets project, it is important to highlight that they suffer mainly from 3 types of fungus: Apple Scab (Venturia Inaequalis), Alternaria Mali and Podosphaera leucotricha. All of them are difficult to treat in a curative stage, so prevention is of utmost importance. They all appear with the right combination of humidity and temperature, and rain generates a spreading effect.
To combat these fungus, the digital platform with logged data, together with prediction algorithms based on the RIMpro Apple Scab Prediction Model, suggests when to apply the spraying. Typically, the prediction stage is accurate within just 3 days before the rain, so usually the farmers use this time for spraying. Given the area of the orchard, and the limitation of working hours, it takes 7 tractors working 9 hours per day to cover the whole orchard with these prediction times. They also apply the spraying after it rains.
In this context, the pilot consists of automatizing a tractor in order to perform the spraying operations autonomously, together with sensors to detect the health of the plant to apply the right kind and amount of fungicides, at the right time, in the right place. Also, Smart Droplets will build a digital twin of the farm, with AI based prediction models implemented in order to improve the time and precision of the prediction algorithms, with the final goal of decreasing the appearance of these fungus.
Given the descripted operative, the apple orchard waste is, in average, of about 8% of the total production per year (apples affected by fungus).
With the help of more precise AI based prediction algorithms, together with autonomous tractors equipped with sensors to detect the health of the plant in order to apply the right kind and amount of fungicides, at the right time, in the right place, Smart Droplets project expects to decrease the waste.
Also, as the autonomous tractors are able to work 24/7, and they will be able to apply the right amount of fungicides, there are also expectations to decrease the operative expenses and fungicide expenses in a great manner.
Our partner Eurecat hosted an engaging Open Day for 40 local farmers and stakeholders. The event featured live demonstrations of the autonomous tractor, providing attendees with their first look at how our technology tackles critical challenges like targeted application for fungal disease management in high-value crops.
An integration week took place at SERRATER SL premises in Spain, where partners Eurecat, the Agricultural University of Athens, and AGROMA worked together to integrate all physical components: the retrofitted tractor, navigation unit, perception unit, and advanced DIS sprayer. Using simulated inputs from the Smart Droplets Digital Twin, the tractor successfully carried out realistic spraying missions—navigating, stopping, and spraying precisely as instructed without human intervention. This validated the system's operational performance in a complex orchard environment.
We are now preparing for our next demonstration in Spain, where the fully integrated system will showcase its advanced capabilities in targeted, intelligent spraying. Official dates and registration details will be announced shortly.
