Author profile picture

When I say, tomato-picking robot, what do you see?

A robotic arm? Something that gently grabs and picks the tomato? Maybe also the (autonomous) vehicle with that robot arm…. In short, hardware, something tangible, or as we also say, “something that hurts when you drop it on your toes.”

However, a robot is much more than just hardware. To explain this, we first go back to a commonly used definition of a robot:
“A non-biological, physical system that sees the world around it and reacts to it.”
In this, we break it down into sense, think and act.

The first step, sense, is about observing the environment. For this purpose, sensors are used that can see, smell, feel, or hear the environment. Depending on the application, it could be a simple camera that can only distinguish between light and dark. But it could also be one with a lidar system whereby a very precise 3D representation of the fruit and its environment can be made. The data from these sensors is stored so that it can be used in the next step.

The second step, think, is about making decisions based on what is observed in the environment. The data collected in the first step is analyzed and a decision is made on what should be done. In the example of the tomato picking robot we have to determine whether the tomato is ripe and needs to be picked. Models and artificial intelligence play an important role in this.

In step three, act, an action is performed. Hardware often is an important part of this. The fruit is picked by gripping it with a gripper, or using suction to suck it up using a vacuum cleaner-like system, or, in the case of vine tomatoes, snipping the stems.

Autonomous harvesting vehicle

So for these three steps, a lot of technology is needed. A robot is made up of many different parts. These vary from the gripper’s hardware, materials, and electrical engineering, to the software that allows the harvesting vehicle to drive autonomously using position determination, data storage, sensors, and AI.

But it does not stop there either. In order to develop the models that determine what must be done, knowledge from the growers is needed. Which tomato is ripe and which one should be picked the next day? Furthermore, the data must of course be stored securely. A picked fruit has to be taken to the shed for sorting and packing. The origin of the fruit should be stored and passed on further up the chain. Perhaps we could make use of blockchain here to make this more transparent.

Climate computer

To complete the picture, it is important that data from the cultivation itself is also included. After all, the temperature, amount of light, and all actions during cultivation affect the fruit’s shelf life and quality. To do this, the harvest data should be combined with, for example, that of the climate control computer and sensors that are installed in the greenhouse. During the following stage, the temperature, humidity levels in the truck, and the duration of the transport are crucial for the quality. With those details about the transport, the shelf life, and the traceability of the product, we come to understanding the reason why we develop all these systems. As the Netherlands, we are the second-largest food exporter in the world. It’s still bizarre to realize that we manage to do that in such a small country!

Produce for our own region

If we want to continue doing that over the coming decades, our system will have to change. We will have to make more use of the available technology. This change has already started! Where we once used to send planes full of peppers all over the world, we now produce much more for ourselves and our surrounding regions in Europe, a radius of about 500 kilometers around the Netherlands. In addition, we provide knowledge and technology to the rest of the world so that food can be produced in a sustainable way around the globe.

We will have to work together even more to maintain this leading role. We have a lot of green knowledge. As well as companies that have been developing technology that has been selling all over the world for decades. However, technological developments are now going so fast that we are no longer capable of developing them by ourselves. This requires collaboration with universities, developers from other sectors, the government, but also companies in the horticultural sector. We need to work with companies worldwide that are developing technology in order to achieve this. We are not just talking about “simple” automation but about digital transformation.

So let’s develop products and services together by making use of the knowledge that’s available around the world, both inside and outside of the horticultural sector. Instead of all just inventing our own wheels with the doors closed!

Would you like to know more about the application of robots in horticulture? Read also: And it can even pick tomatoes, Vegetables of the future adapted for picking robots and take a look at

About this column

In a weekly column, written alternately by Wendy van Ierschot, Eveline van Zeeland, Eugene Franken, Jan Wouters, Katleen Gabriels, Mary Fiers, and Hans Helsloot, Innovation Origins tries to find out what the future will look like. These columnists, occasionally supplemented with guest bloggers, are all working in their own way on solutions for the problems of our time. So tomorrow will be good. Here are all the previous articles.