The Eurecat technology centre, the Basque Government’s Department of Education through the Basque Centre for Applied Research and Innovation for Vocational Education and Training and Enkitek have developed a smart robot harnessing computer vision and artificial intelligence for harvesting delicate fruit such as strawberries. The robot was unveiled at the Expo AgriTech 2024 fair specialising in agricultural technologies held from 26 to 28 November in Malaga.

Developed as part of the HarvBot (HARVesting coBOT) project, the robot is equipped with computer vision and artificial intelligence technologies to detect the fruit and has a data system to monitor the state of the crop and the harvesting process.

The project “has designed and tested new 3D printed components to customise the robot to identify and pick the fruit along with algorithms for detection and handling tasks,” says Óscar Palacín, a researcher in the Robotics and Automation Unit at the Eurecat technology centre.

“The current solution has been validated with strawberries but can also be tweaked for various types of crops including tomatoes, peppers, cucumbers and stone fruits.”

The robotic solution, which has been developed in response to a market need pinpointed by Tknika, the Basque Centre for Applied Research and Innovation for Vocational Education and Training as the sponsor and driving force behind the project, has been validated in a hydroponic crop, i.e. without soil, with strawberries at Tknika’s own facilities in Errenteria, Guipúzcoa, and used as a pilot scheme for Enkitek.

“Applying software agnostic to the hardware used, the main building block in this project, is crucial to introducing automation in this emerging sector, providing access to the growing supply of low-cost robots and the chance to become scalable businesses,” comments Victor Canton Ferrer, Enkitek’s CEO and co-founder.

“So far we have achieved extremely successful outcomes,” adds Pili Alonso, executive director at Tknika. “This kind of project encourages and promotes training and competitiveness in applied innovation knowledge transfer.”

The robot has some new innovations compared to the first developments since “a database has been added for real-time monitoring of the picking process coupled with a graphic interface for its display,” points out Gennaro Scaratti, a researcher in Eurecat’s Robotics and Automation Unit.

“These new features will make it possible to seamlessly embed the solution in crop management systems for real-time monitoring, analysis and decision making.”

The HarvBot project has been run with the backing of ACCIÓ, the Government of Catalonia’s Agency for Business Competitiveness, under the Digital Innovation Hub of Catalonia.

“Robotic solutions such as this one help to meet the growing need of agri-food companies to optimise in-field tasks given labour shortages,” notes Carol Benedí, Food Market business developer at Eurecat. “Plus they generate a crop monitoring database which enables more sustainable and resilient management in response to climate change.”