Robots in strawberry fields: How AI automates harvesting and supports farmers

For larger farms such as Erdbeerhof Glantz, ensuring a reliable harvest under these conditions is particularly challenging.

SHIVAA: Mobile harvesting robot for open fields

As part of the RoLand project, the Robotics Innovation Center of DFKI, together with the Research and Transfer Center Smart Systems at HAW Hamburg, Othmerding Maschinenbau GmbH & Co. KG, and the associated partner Erdbeerhof Glantz, developed the mobile harvesting robot SHIVAA. It was specifically designed for use in open fields – an environment that has so far been particularly challenging for machines.

Sensors and AI ensure quality around the clock

Here’s how the robot works: Using a 3D stereo camera, SHIVAA autonomously follows a row of strawberries. Additional cameras, including multispectral sensors, capture the position, shape, and size of each berry. A detector trained with machine learning then reliably identifies which strawberries are ripe for harvest. SHIVAA picks these fruits directly from the plant using a specially designed robotic arm. The arm approaches the fruit at high speed, attaches it with a specially developed gripping mechanism, and gently encloses it before placing it in the harvesting crate carried on the robot.

A major advantage of this technology is its ability to operate around the clock: Equipped with a lighting system, SHIVAA can harvest at night or in the early morning when temperatures are low. This protects the delicate fruits, reduces losses, and ensures consistently high quality.

Seamless integration into farm operations

The robot is designed to specifically relieve human workers. It can compensate for short-term labor shortages and work directly alongside people in the field. In the future, systems like SHIVAA could be integrated into farms of all sizes – from small family-run operations to large-scale plantations.

The technology also stands out for its scalability: deploying multiple robots simultaneously makes the investment particularly cost-effective. Additional benefits could come from potential upgrades, such as modules for UV light-based fungus prevention or mechanical weed control. In this way, the robot could also help reduce the use of crop protection chemicals.