H2020 GALIRUMI Project: Galileo-assisted robot to combat the Rumex obtusifolius weed and increase the profitability and sustainability of dairy farming

Satellite navigation enables weed control without herbicides. Satellite-guided autonomous robots can perform weeding tasks with precision, helping farmers reduce herbicide use and eliminating the need for backbreaking manual labor. While agriculture provides essential food and raw materials, the environmental impact remains unacceptably high. This has led to concerted efforts to make the sector more sustainable. "In all aspects of agriculture, controlling weeds is a major issue," notes Esther López, GALIRUMI project coordinator at ACORDE Technologies in Spain. "However, weed control with herbicides is harmful to the environment, while manual weed control involves backbreaking work and high labor costs."

The EU-funded GALIRUMI project sought to develop an innovative new weeding method that overcomes both of these drawbacks. The project's solution was to apply robotics to provide herbicide-free weed control on dairy farms, particularly for broadleaf sorrel. "Robotic weeding not only eliminates the use of herbicides but also reduces farmworkers' exposure to them," says López. "This approach can also contribute to the growth of organic farming." Using precise satellite navigation.

The project team began by examining various innovative technologies for weed detection and degradation, as well as for autonomous vehicles. All of this was based on the precise navigation provided by the European Global Navigation Satellite System (EGNSS). The technologies investigated included computer vision for weed detection, as well as laser and electric weeding systems. The project consortium included experts in weed and dairy research, as well as technology specialists. “We involved end-users from the beginning,” notes López. “They provided us with specific requirements and operational scenarios, which contributed to the design and development of the robotic systems.” The completed robots incorporate EGNSS modules with dual-antenna configurations, camera systems for weed detection, and a central processing unit. The dual EGNSS antenna provides precise directional information to the robot’s navigation module.

An economic feasibility assessment of the robotic weeding solution was also conducted. Autonomous Weeding Demonstrations The robot prototypes were subsequently tested on dairy farms, focusing on broadleaf weeds. Two types of robotic weeding units were developed: one for laser weeding and the other for electrocution weeding. "We were able to demonstrate how the robot follows a predefined route through the field," explains López. "When it detects a specific weed, the robot can deviate and navigate to the coordinates of the detected plant." The robot stops over the weed and activates the laser or electrocution function. Upon completion of weeding, the robot returns to the predefined route and searches for more plants. By assisting farmers with weeding tasks, López believes the laser robot is closer to commercialization.

The weeding unit is independent of the vehicle, and the vehicle is light enough not to damage the lawn along the way. More work will be needed to commercialize the electric weeder. "We are still identifying some commercial aspects and trying to better understand the economic concerns of farmers and future users," adds López. "However, we have been able to demonstrate that an autonomous weeding robot, with the support of EGNSS, can assist farmers in their weeding tasks." In this way, López believes the project can significantly contribute to organic production, eliminating the need for herbicides and labor, and generally contributing to environmental protection.