Skip to main content
Farma Ráječek Jahody.

AI helps to monitor and harvest strawberries

Sustainable crop growing with fewer sources and without using chemical products is the future trend in agriculture. Robots play a crucial role in this scenario. An example is Fravebot, a robot that monitors and harvests strawberries at Ráječek family farm in Brno, Czech Republic.

The path of a strawberry from the tiny seed in the greenhouse to the sweet fruit on the supermarket shelf is long and winding. It involves many steps and a lot of manual labor. Or rather it did. In their approach of revolutionizing strawberry harvesting with robotics Ráječek farm uses two types of robots to make farming strawberries easier and more sustainable:

  • The Fravebot Scout for monitoring plant health and ripeness
  • The Fravebot Harvestor for picking the fruit

"Dividing work tasks between two robot types is beneficial in many aspects," explains Vratislav Beneš, Chief Constructor at Fravebot and the company's founder. "The Scouts, whose primary task is to monitor plants, move around the greenhouses faster, and if they were to do some other work simultaneously, it would delay them. It is the other way round – working robots work effectively in cases where they drive out only where they are needed."

Fravebot stands for Fruit and Vegetable Robot, but internally the robot is affectionately called Franta. Students of the Brno University of Technology and the Mendel University in Brno were also involved in its development.

Digital twin in agriculture

On top of the robots a digital twin environment by NVIDIA is key on the Ráječek farm. It helps develop, simulate, and deploy an AI that is meant to be the robots brain.

Using the digital twin of plants and fruits in the NVIDIA Omniverse the robots are able to detect diseases and pests is the real world. The interface between real and digital world comes from the open digital business platform Siemens Xcelerator.

As today's simulation tools can also simulate physical laws (e.g., the weight of strawberries), the robot's neural networks can be trained in advance without the need to move around the real environment.

This approach significantly accelerates the robot development process and reduces costs.

"I think that is the main path to follow: not to focus merely on the robotic part but also analytics and the user interface for plant growers," emphasises Matěj Sklenář, co-owner of Ráječek farm and Chief Agronomist of Hydroponic Cultivation.

Thanks to a technology library robot programming and control is easier than ever eliminating the need to have two programming environments. Which means the combination between the real and the digital worlds is closer and simpler than ever.

December 2023
Contact: Radek Hofírek

Farma Ráječek Jahody.

The Fravebot Scout monitors plant health and ripeness.

Showcasing Siemens technology inside, a robotic unit moves on rails for ideal navigation.

The Fravebot Harvestor picks the fruit.

The robots are equipped with Simatic S7-1500 controller and other Siemens technologies.

The robots are equipped with Simatic S7-1500 controller and other Siemens technologies.

Colorful hot air balloons floating in a clear blue sky above a green landscape

The team of Siemens, Ráječek farm and Fravebot have revolutionized plant monitoring and harvesting (from left): Radek Hofírek, Product Manager, Siemens Digital Industries, Czech Republic; Matěj Sklenář, Co-owner of Ráječek Farm and Chief Agronomist of Hydroponic Cultivation; and Vratislav Beneš, Founder and CEO of Fravebot.