Inventing the agriculture of the future

Summary

Drones to monitor plots of land, peer-to-peer tractor rentals, sensors in fields, smart cows, and much, much more: entrepreneurs are vying to outdo each other with ingenious ideas for inventing new agricultural practices. Let’s take a closer look at some of the many innovative initiatives around the world combining tech and sourcing.

25Août.
0

Real-time agriculture

For several years now, agricultural professions have been undergoing a veritable digital revolution, with the arrival of a generation of smart tools that let growers optimize their farms’ management and monitor a whole series of key components for the health of their herds and the yields produced by their fields, all in real time. The stakes are high, because the UN has estimated that agricultural production will need to increase by 70% by the year 2050 if it is to meet global demand. Below is a sampling of some of the existing solutions.

Smart cows

The first smart cow project was born in Japan, at the initiative of Fujitsu, a company specialized in the IoT[1]. The project’s goal was to equip cows with motion sensors. Because ovulating heifers are more active and move more quickly, the data collected are analyzed in real time to determine the perfect time for inseminating each of them. A similar experiment is being conducted in England, led by Zoe Barker, a PhD and lecturer at Writtle College. The goal is to demonstrate that a cow’s position can provide precious information on its state of health and wellness. This tool is able to identify animals that are tired and systematically lie down apart from the herd and then inform the teams so they can treat them as quickly as possible, which can help to contain any contagion. “Some diseases are particularly difficult to detect during routine check-ups by veterinarians,” Zoe Barker explains, “such as mastitis, which costs English farmers more than ₤100 million [more than $130 million] each year.” Based on that data, Zoe Barker and her team plan to develop a behavioral model that will allow them to build a prevention tool for livestock farmers.

Diagnostic tools

When applied to the monitoring of fields, the “Internet of Agricultural Things” is incredibly promising.

The PulsePod, for example, developed by the company Arable, is a pocket instrument worth $500 that measures a field’s water needs, based on the surface area of the canopy[2] and recorded rainfall. This small, portable lab also takes other factors into account to detect environmental stressers like air pollution and insect invasions. The encrypted data is transferred over an API[3] that gives farmers real-time reports on the state of their fields and, eventually, allows them to optimize their water consumption and reduce the use of pesticides.

To collect the data, drones are on the front lines in the United States, a country endowed with gigantic farms and legislation that is favorable to the commercial use of light drones. In 2015, an estimated $320 million were invested in start-ups developing drones or robotics for agriculture in the United States. These include the teams at Precision Hawk and the French-based Sensefly, both of which have strong positions on the US market. These are joined by Marvx, a start-up that utilizes drones owned by its partner companies and private individuals to obtain images that it then analyzes on its platform for its clients.

Once gathered, this growing quantity of data then needs to be analyzed. To help farmers with their diagnostics, the company Prospera, a specialist in agricultural big data, trained its algorithms on millions of snapshots of plants to teach them how to distinguish between healthy leaves and infected leaves. Every day, Prospera runs its detection software on 200,000 images of greenhouses, captured throughout Europe, Israel and Mexico. Its visual recognition system analyzes the light spectrum of each leaf, to assess its maturity, detect any diseases and estimate insect population numbers. In this way, threats can be detected sooner so that action can quickly be taken to deal with them. “Farmers who are accustomed to making decisions based on instinct will be able to look to reliable, up-to-date data,” exults Daniel Koppel, CEO and co-founder of Prospera. “Our system has already optimized our clients’ yields by an average of 30%.” The company, which is currently working on greenhouse crops, will soon be launching its services in outdoor fields.

Peer-to-peer agriculture

Digital technology is also affecting agricultural hardware through co-farming platforms like WeFarmUp, a pioneering French company that launched the first peer-to-peer website for rentals of agricultural equipment: plows, seed drills, mechanical weeders, soil equipment, handling equipment, spreading equipment, and more. “These expensive machines are only used a few days out of the year and spend the rest of the time tucked away in hangars,” explains Laurent Bernède, founder of this young company that aims to revitalize local mutual support practices for agriculture through this digital cooperation. All rentals are covered by an insurance contract and a security deposit, which give the owner the necessary protections. Each farmer is paid 85% of the price of the rental, and WeFarmUp receives a 15% commission. Today, mechanization expenses gobble up more than one-third of the revenue of agricultural concerns. This pooling of expenses not only allows the equipment’s owners to generate a cash flow and gain a return on their investments more quickly, but it also lets new farmers just starting out to obtain the equipment they need, all within a 12-mile radius. In France alone, WeFarmUp estimates there are 500,000 farmers whose equipment could be rented via its online platform.
Soon enough, all farming activities will be able to use the convergence between big data, artificial intelligence and robotics to optimize their activities. And while agriculture is still far from being an exact science, this marks another step forward toward improved resource management.

Photo ©Precision Hawk

[1] The Internet of Things
[2] The canopy is the part of a field’s vegetation that is directly affected by solar radiation.
[3] An API, or application programming interface, is a computer interface through which software offers services to other software.