Russia Develops AI System to Cut Herbicide Use in Agriculture

A new AI-powered technology aims to reduce chemical use by up to 40% while easing the environmental impact on soil.

Researchers at Don State Technical University in Russia are developing an artificial intelligence system to help farmers combat weeds more efficiently and at lower cost. The project, called AgroVision, uses the YOLOv11 neural network architecture to precisely detect the location and type of weeds directly in the field, RBC reports.

This precision application approach will allow herbicides to be applied only where needed, reducing their use by 30–40% and minimizing chemical impact on the soil. In the future, the system will be able to identify specific weed species, further improving treatment efficiency.

The project is expected to be completed within a year, with tasks including data preparation, prototype creation, and adaptation for commercial deployment. Implementation in the Rostov region’s agricultural sector could be a major step in the digital transformation of Russian farming.

Previously, the university announced work on an AI platform for blood analysis to speed up medical diagnostics. Together, these initiatives reflect the region’s drive to integrate advanced technologies into critical areas of life.