Wetenschap - 9 februari 2017

Tractors and drones advise farmers

Albert Sikkema

Smart farming, making use of ICT in agriculture, is on the rise, and unstoppable. But how can we make sure all the data it produces actually benefit farmers and consumers? WUR researcher Sjaak Wolfert is working on answering this question.

illustration Shutterstock

New sensors, meters and software systems for planning, implementing and analysing our food production are following each other onto the market in quick succession. Robots are taking over some of the work on the farm, and big data on prices, market conditions and consumer behaviour exercise more and more influence on chain management. This is going to lead to a shift in power relations in food chains, expects Sjaak Wolfert, research coordinator of the EU project Internet of Food and Farm, which started in January and forms part of the EU programme Internet of Things.

Wolfert, who works at Wageningen Economic Research, is studying the potential and uses of digital production in the agriculture sector. He and his European partners are mainly developing technology for on and around the farm, but they are also interested in logistics and consumers in the food chain. The total research budget for his project is 30 million euros.

Targeted spraying

The starting point for Internet of Food and farm is the idea that smart ICT use can make the food sector more efficient and sustainable. You could take the example of drones which can map out spreading pests so precisely that farmers can target them with spraying; or that of chips which monitor the freshness of vegetables en route from the grower to the consumer. And then there are possibilities such as apps in which consumers can fill in their food preferences and then, when they scan a product in the supermarket, be told whether they will like it.

The project focuses on technology which is not yet on the market. The aim is to get that technology adopted across the EU, says Wolfert. His project facilitates a successful launch, creating the right conditions and finding partners. Sometimes a better communication infrastructure might be needed. Since Wi-Fi or 4G are not available everywhere in the EU, the project partners are working on networks of their own, such as LoRa or SigFox, which can communicate over long distances using little energy. In many other cases, there is a lot of information technology already in place, and the main issue is to connect up data, explains Wolfert. For example, weather forecasts and information obtained from satellites and tractors can be linked with data on cows. ‘The emphasis in our project lies on analysing and combining information in order to arrive at good advice for farmers and consumers.’

In the related Dutch project Data-Fair, a joint venture with insurance company Achmea, a drone films a crop and a programme interprets the film in terms of the crop’s commercial value. This is useful information if, for instance, the crop is damaged by a severe hailstorm. After the storm the drone can fly over the field again, and the insurance company can calculate and pay out the damages.

As long as the farmer wants to share the data with the insurance company, that is, adds Wolfert. Because the example raises the question: who do these data belong to? This is definitely not just an academic question. Who owns the information from milking robots? These robots amass raw data on cows which the farmer cannot make use of. Lely, a company that makes milking robots, says: this information belongs to us. Does it? From a legal perspective, it is not possible to establish ownership of data, says Wolfert. ‘So you have to make agreements between the farmer and the provider on the use of the data.’

Google and Monsanto

There are now a lot of companies active in the field of digital data processing and smart farming. And nearly all of them have their own platform and service. Lely offers agrarian services using milking robots; tractor manufacturer John Deere using tractors. Wolfert: ‘They are all starting their own platforms and apps, but often these don’t work in combination and farmers don’t want to use ten different platforms. So we are also working on coordinating processes.’

The government needs to stay alert with regard to the way smart farming is developing, says the researcher. ‘The biggest companies in the world are now data companies. You can earn a lot of money and gain power with data about other people, as we have seen from Facebook. We fell for that one, you could say. The question is whether and how you can prevent that in the agrofood sector. Take Monsanto’s model FieldScripts. The farmer and Monsanto provide data on the fields and the farm, and Monsanto then gives advice on which variety the farmer should grow. And of course, that will be Monsanto seed. This is an example of ‘lock-in technology’ in which the company decides on the farmer’s choices. There is an alternative for American farmers: the Farmers Business Network. That is the property of farmers, but beware: Google plays a role there too. So be alert to who is developing the smart farming.’


In order to stay independent of these large seed and media companies, in the previous EU project, Future Internet, Wolfert worked a lot with small and medium companies and start-ups in Europe. In the last phase of this project, Wolfert had about 50 European start-ups with an interesting application. They were all given 50,000 euros to create a business plan and pitch their idea to venture capitalists. Several of these companies were able to implement their plans with external capital. This successful approach put Wolfert and his fellow researchers on the radar in Brussels. And partly as a consequence of that, they can now do further research over the next four years under the auspices of Internet of Food and Farm.

The main aim of this further research is to broaden the application of ICT, explains Wolfert. ‘Imagine: I’m driving my machine across a field and the camera on the machine registers that there are a lot of weeds. In our previous project we managed to convert the camera images directly into a decisionmaking model that decides whether the machine should start weeding. In this new project we are going to see whether we can also use the camera data for other purposes such as harvest forecasting.’

Precision farming

Crop farmer Jacob van den Borne from Reusel in Brabant uses the latest ICT techniques, monitoring his crops with drones and using tractors full of sensors. WUR researchers see his approach as an example of smart farming. Van den Borne himself calls this way of working ‘precision farming’. The technology enables him to work more efficiently and thus to save money, he explains in a video.