METBOTS - Metabolomic robots with self-learning artificial intelligence for precision agriculture

FINISHED (September, 2021)

The development of a robotic platform - for precision metabolic monitoring in viticultural production, generating georeferenced images of vine metabolism, grape and soil composition, through spectroscopy techniques - is the aim of this project. These data will be correlated and interpreted by a self-learning artificial intelligence system to assist in the precision viticulture context.

The MetBots project is divided into three parts:

• Robot and sensors - development of hardware and self-adaptive arm system for collecting spectra on the vine, grapes and soil, vision system, machine for navigation and positioning, estimation of vegetative indexes and production volume;
• Cloud system - houses the project databases, spectra processing and self-learning artificial intelligence for accurate quantification of metabolites by spectroscopy and data analysis system;
• Massive data with information fusion for the management of precision viticulture practices;
• Field validation tests - blind validation tests of the quantifications obtained by spectroscopy vs. certified laboratory, as well as, comparison and analysis of the information collected against prior knowledge of the complex interaction of the experimental field.