Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Bulgaria

European Innovation Council · Running

RoboAIweeder A fully autonomous solar-powered lightweight weeding robot, using AI for plant recognition, precision contact and contactless weeding methods suited for hard soils, hilly terrains and arid climates.

Coordinator
Smart Farm Robotiks, Bulgaria
Runs
1 June 2024 to 30 November 2026
Status
Running
EU funding
€2.4m
Total cost
€3.4m
Partners
1 organisations in 1 countries
Type
European Innovation Council
Grant agreement
101166300

What the project is doing

In the project's own words, from its summary on CORDIS.

Fighting weeds is one of the oldest problems of agriculture. The current solutions include mass use of herbicides, mechanical tilling or manual weeding. Herbicides are poisons which kill not only weeds, pollute the soil, water and bring lasting damage to all living organisms, including health and well-being of humans. Mechanical weeding with standard agri machines (ICE-based) adds to CO2 emissions, damages and compacts soils, and is very imprecise.

And manual weeding is back-breaking, often dangerous, low-paid work for which there are fewer and fewer candidates. Our solution addresses all these environmental, economic and health problems together. It will also strongly support the expansion of organic agriculture in Europe, which is facing much more difficulties with weed pressure, compared to conventional farming. And our innovation is also specifically designed to target the needs of the most climate-vulnerable farmers – those from the semi-mountainous regions of Southern Europe.

We offer a fully automated solution in the form of a 4-wheeled rover, with in-wheel integrated electrical engines, solid X-Y-Z wheel synchronization and front-steering mechanism. It is lightweight and solar-powered for 24-hour working cycle. The robot uses AI based on deep neural networks to spot weeds among desired plants. It can effectively destroy weeds using contact (mechanical) and non-contact (energy beam) methods, depending on both weed size and type, and soil and weather conditions, without creating any fire hazard in the process.

Finally, our robot can self-navigate in and around the fields, which is achieved without costly RTK equipment, by combining multi-GNSS (including Galileo) receivers and cameras with proprietary algorithms. The robot has huge potential to expand its capabilities by adding sensors for soil, weather tracking, etc., and we can grow our offer by supplying vast plant and field data analytics and predictive analysis to farmers and other potential clients.

Topics

Who is involved

OrganisationCityCountryRole
Smart Farm RobotiksSofiaBulgariaCoordinator

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI