Fields of innovation

Home Technologist 05 High-tech farming Fields of innovation

Farmers, technophobic? Hardly. To improve crop yields, reduce costs and move towards more sustainable practices, the agricultural world is turning to such cutting-edge technologies as drones, robots and networked sensors.

Cow in VMS

090Stereotypes are hard to shake, but that of the progress-resistant farmer is just wrong. “Farmers have always embraced innovation,” says François Purseigle, professor of sociology at the École Nationale Supérieure Agronomique in Toulouse. “Since the beginning of the 20th century, technology has emancipated them from some of their most backbreaking work, while also giving them better control over their crops and improving yield.”

The stakes have never been higher. The world must find a way to feed a population that will grow from 7.2 billion today to 9.6 billion in 2050, according to the United Nations. Doing this with minimal environmental impact is also important since agriculture is thought to be responsible for 30 percent of the world’s CO2 emissions. High-tech farming will play a critical role in meeting these challenges.

Drones everywhere?

colza_biomasse

Hyperspectral images taken by drones allow farmers to analyse everything from chlorophyll level to water stress.

The solution could come in part from the sky. Drone use is becoming widespread, thanks to relaxed regulations and a growing number of manufacturers. “We provide agricultural clients with crop-monitoring and decision-making tools,” explains Albert Maas, co-founder of Avular, a Dutch manufacturer founded in 2014. In addition to providing images with 10 times higher resolution than satellite images, “the main advantage of the drone is that it can operate below the cloud cover, making it less dependent on weather conditions and therefore more responsive.” Drones are used to analyse a vast number of parameters including chlorophyll levels, nitrogen requirements, soil reflectivity levels, vegetation density and water stress. “The goal is to move beyond empirical experience and manage the entire agricultural cycle based on a massive collection of data,” says Maas. The information is processed by a series of software tools that map the fields, allowing the farmer to fine-tune supplies of water and fertiliser. Farmers can also observe early signs of disease and apply pesticides in a targeted manner. The downside of drones is still their cost. A satellite image costs €8 per hectare. A drone image can easily cost five times as much; add to this the price of the drone itself, which can run anywhere from €25,000 to €50,000, plus the associated software. These costs are likely to fall, but for many farmers they will remain prohibitive.

Sensors and data harvesting

High-tech farming also relies on exponential growth in the volume of data collected for a wide variety of uses: weather forecasting, irrigation control, alarms (against theft of livestock or material) and animal-health monitoring. As in other areas, the arrival of Big Data brings a host of technical challenges. Sensors deployed in fields and barns must have a high level of autonomy to spare farmers from worrying about power supplies. Transmitting information can be a headache, since open fields do not usually enjoy the same mobile coverage as urban areas. Companies like Sigfox, which specializes in wireless networks and inexpensive connected devices, may soon fill this niche. Based in Toulouse, Sigfox recently raised €100 million from European, American and Asian investors to extend its international development. They could respond to the needs of projects like E-Pasto, a Franco-Spanish device that acts as a virtual livestock enclosure. Animals are equipped with a GPS device and battery; when they get too close to the pasture boundary they receive an electric shock that’s about as strong as the one they would get from a traditional electric fence. The device is especially suitable in mountain areas to protect animals and save money on fences.

Robots making tracks

The French-made Oz, an electrically powered robot for mechanical weeding, runs autonomously for up to four hours. It costs less than €1 per hectare to operate.

The French-made Oz, an electrically powered robot for mechanical weeding, runs autonomously for up to four hours. It costs less than €1 per hectare to operate.

Another promising field is robotics. Here again, small and medium-sized companies are finding inventive ways to widen the range of applications for the machines that are still mainly used in livestock; traditionally, 87 percent of agricultural robotics has been used for milking. Milking machines are expensive – about €300,000 for 120 cows – but above a certain herd size they’re indispensable because of their economies of scale. Other applications either ease the backbreaking labour of farming or speed up operations. Agrobot, a Spanish company, developed a robot that can pick strawberries; it is already being used widely in California. “It takes only five minutes to learn how to operate the machine, and it produces significant savings,” explains Juan Bravo, founder and CEO. “Not only is the harvest faster and easier, but the robot makes up for the lack of available manpower.” John Deere, the U.S. agricultural machinery giant, is taking the same approach; they recently released automatic seeders that can work at nearly 20 kilometers per hour.

Oz and Cosi, robots made by French manufacturer Naïo Technologies, move automatically and silently through fields, hoeing the soil and removing weeds. “Their automatic navigation system allows them to cross rows without crushing crops and remove weeds that they recognize through an algorithm,” explains company co-founder Aymeric Barthes. This reduces or eliminates the need for pesticides and manual weeding. It’s another small example of how technological progress is bringing the goal of intensive yet sustainable agriculture closer to reality.

By Jean-Christophe Piot

SIMILAR ARTICLES

Robots vs Jobs

The looming confrontation
Technologist Issue 13

A universal basic income would mitigate the negative effects of automation. But it might be more effective if combined with…
Robert West, researcher at EPFLRobert West, researcher at EPFL

A Swiss researcher has created a system that scans Wikipedia for important articles that are missing in other languages. This project…
pepper_robot

Humanoid robots are finally learning to charm us. Can French-founded SoftBank Robotics stay ahead of the competition after a decade…