Researchers create prototype drones to predict wine production3 min read
Researchers at the University of Porto will use artificial intelligence and create micro-drones, tomographs for grapevines and applications for smartphones to predict the productivity of vines in relation to the climate in a project funded by the PRR.
The scientific trials of the Wine4cast project will take place over the next three years in the Demarcated Regions of Douro and Dão, North and Center of Portugal, and the creation of models to predict vine productivity using artificial intelligence, sensors and algorithms will be funded with an amount of about one million euros from the Recovery and Resilience Plan (RRP).
“This is very important for Portugal, this productivity forecast, because in addition to the interannual oscillation that is characteristic of the vineyard, in Portugal we are very influenced by the Mediterranean climate and these oscillations are huge,” said today to the Lusa agency Mário Cunha, the project coordinator, researcher and professor at the Faculty of Science, University of Porto (FCUP).
Using artificial intelligence to obtain more accurate vineyard productivity forecasting models is one of the project’s missions, which will then make the scientific findings available to wine growers.
“From the technological point of view, the research project will use tomographies that will allow us to understand how the plant’s organs are, and this will allow us to make an early prediction, about 17 months before the harvest,” explains Mário Cunha.
The tomographies will allow us to analyze the “fertility of the hibernating lumps”, because the vine has a two-year cycle.
“We will be able to perceive inside the organs of the plant [vine], what will be the potential of the production that will occur 17 months later. This is a new technology that we will develop from scratch, with technological partners in the project,” revealed the expert.
Another prediction model is micro-drones that are only 10 centimeters in size, to fly over and inspect the grapevine, finding out with this technology how many bunches and grape berries there are.
Smartphones will be another technology used to support the prediction of wine productivity.
Mário Cunha reveals that they will develop an application for smartphones, allowing winemakers to digitally inspect and map the vineyard and count bunches and berries.
The productivity forecast will also allow “fraud control” in the sector, particularly in years when there is low productivity, and can avoid the flow of transnational or even transcontinental grapes or wine, as well as be able to strengthen policing measures and the issue of insurance premiums.
“More and more companies want to take out insurance and insurers need to have information to understand what the forecasts are to set an insurance premium,” explained the expert.
In the specific case of Port Wine, the Wine4cast project will help to know in a more timely manner how much brandy needs to be imported, since with the production forecast, it will be possible to avoid buying too much or too little brandy.
“It will be possible to forecast the number of harvests, for example, from 2025 to 2050, as well as to know the number of harvests above average and what the number of harvests and productivity levels will be.
Wine4cast is a consortium with FCUP, in partnership with the Institute for Systems Engineering and Computers and Technology and Science (INESC TEC) and partners from the wine sector at a national level.
The project is included in the “Agriculture 4.0 and Sustainable Territories” initiative of the PRR’s Innovation Agenda for Agriculture 2030, a national implementation program, with an execution period until 2026, which aims to implement a set of European reforms and investments to help Portugal restore economic growth after the pandemic.