This paper presents a multi-robot system for the autonomous detection and treatment of pests. The
system comprises two inspection robots, based on commercial electric vehicles, and two treatment
robots equipped with electric sprayers. All four robots are fitted with cameras and global navigation
satellite system (GNSS) receivers and operate under a cloud-based mission management system.
The robots function decoupled, enabling autonomous navigation, real-time pest detection using
convolutional neural network (CNN) models, and precision treatment application. The system was
validated in Botrytis detection in white grape vineyards, demonstrating accurate path following,
continuous image acquisition with an average 20 % overlap, and 95 % detection accuracy with a
30 % false positive rate., This work has been funded by the EU FlexiGroBots project (https://cordis.europa.eu/project/ id/101017111), Peer reviewed