Weeding wild vegetable species growing in vineyards and orchards farmland is a costly process, which needs to be repeated periodically, being typically done by mechanical and chemical methods. Plows and cutters are used to remove herbs between plant rows and herbicides are applied on the line between plant feet, in order to kill the weeds. Both methods are considered very aggressive for the cultures. Moreover, chemicals remain in the environment and may contaminate water lines, being harmful both to the environment in general and to the final consumer. These processes also entail a significant economic impact, representing, in the vineyards', 20 to 35% of total working time, with typical costs per hectare ranging from 80€/ha up to 380€/ha.
The use of animals for controlling weeds is an old method that has been successfully tested in various regions, reducing the environmental impact and providing land fertilization. However, it cannot be used throughout the year, since animals tend to feed on the lower branches of plants and on fruits, affecting the production.
This project aims to develop an IoT based system, able to control animal posture, limiting their ability to access branches and vine fruits, and to deploy virtual fences to control animal feeding areas. In this way, it will be possible to use herds of sheep to make the weeding of vineyards. This approach is disruptive compared to approaches based on electrification of the rows of grapevines, presenting several advantages:
- Reduced installation costs: the system includes devices coupled to animals and portable RF beacons that define the grazing area.
- Flexibility: the system can be quickly and easily moved between different vineyards, and other types of crops.
- Veterinary monitoring of herd status: the system allows the collection of mobility rates that allow monitoring the animal behavior, with potential interest from the herd health point of view.
- Security: real-time herd supervision allows detection of abnormal situations, such as attacks from predators.