Robots in the dirt: How farming is going digital and why it must

Key takeaways

  • The world’s population is expected to hit 11 billion people by 2100, but land is scarce and farm yields flatlining.
  • Farms must be more efficient, produce more and do less environmental damage.
  • AI, robotics, drones and IoT are just some of the technologies revolutionising agriculture, often in unexpected ways.

Farming is one of the world’s oldest occupations. For tens of thousands of years humans have been cultivating crops and tending to animals to produce our food. Yet while the sophistication of farming systems has increased, it is still not uncommon to see farmers writing weather and crop notes on the dust covered windshields of their vehicles.

Increasingly, agriculture will have to make real progress into the digital era as the world’s population grows, moves further into urban areas and is forced to put a premium on environmentally sustainable practices. As dirt and data intersect, digital will provide solutions to some of the biggest challenges the agriculture industry faces. Here are just a few.

Enhancing farming with digital

Like all industries, farming will increasingly face a new breed of businesses with lower overheads and leaner business models. Efficiency will be a key goal that digital solutions are poised to deliver, particularly when it comes to agtech, precision farming and the ability to make more informed decisions.

End-to-end farm management platforms are an example of this, allowing farmers to merge new and historical data – such as yield history, weather forecasts, crop treatment records or cattle breeding histories – and make choices specific to their environments.* Whether to move stock to a particular paddock, or put down fertiliser becomes an informed decision that saves time and money.

Drones, multispectral and satellite imagery are also providing valuable information at a glance – for instance, whether a particular area or crop is unrecoverable and not worth spending further money on (eg. fertiliser, pesticides). In a similar vein, the Internet of Things, the connected network of sensors and internet-enabled technology, is making light work of mundane but critical farm maintenance. Being able to check the state of fences, machines, levels and health of water tanks, or whether a gate has been shut without manual inspection is just the start. Over time, AI and machine learning will make these operational decisions autonomously.

Other IoT solutions are helping get better production value. Livestock solutions such as Cowlar, a smart collar to track dairy cows, promises to track cow health reducing the the need for antibiotics and increasing milk production.1 Kenya’s iCow is helping farmers by providing an SMS/app service to guide them through best practice management of cow fertility cycles.2

Population and yield

On a macro level too, farming needs to be more productive. By the year 2100, the population of the world is projected to reach 11.2 billion.3 That increase, up from today’s 7.6 billion, will require a lot more food than is currently generated. Additionally, by just 2050, two thirds of the world’s population is forecast to live in cities, up from half now, meaning food will need to travel more to get to those eating it.4

As a species, we are already using 71% of the world’s habitable land, and 50% of that is agriculture (about 51 million km2). Most of that in turn is taken up by livestock. While moving away from meat and dairy consumption would change this balance, and release 3.1 billion hectares of farmland,5 any such change in diet would take time. Lab-grown meat may help in this regard, if it proves palatable, but in the meantime, smarter farming of crops must provide a solution.

Infographic on food allocation

Without more land, crops must provide greater yield. The use of modern fertilisers and pesticides, coupled with farm technology such as irrigation, has already dramatically intensified the yield per unit of land.6 

One of the greatest current barriers to improving the amount of food grown is in the soil – not, as it once was in the quality of the soil, but in the density of it. Industrial farm machinery is big, and heavy. The use of ploughs, tractors, harvesters and sprayers compacts soil, making it harder for crops to develop healthy root systems, and thus less likely to grow to their full potential.

Enter the robots

Robots are increasingly being trialled in agricultural settings, from milking cows, feeding chickens and harvesting fruit** but they are also addressing issues such as soil compaction. Companies like Britain’s Small Robot Company and Earth Rover, are producing small, comparatively light, robots that can do the job of big machines without the traditionally weighty downsides.*** These robots, using GPS and Lidar to orient themselves in fields, autonomously roam crops while doing a variety of jobs, potentially planting, tending and harvesting. Some for instance, take hi-res and infrared imagery to assess individual plant health.7

Such robots do more than automate existing farms – they also enable new ones. Urban farming, previously uneconomical due to compact fields being incompatible with large machinery, becomes possible. Iron Ox in the US, for example, is trialling indoor hydroponic farming using robots to tend and harvest produce.8

As a recent strategy+business piece points out, there are many benefits to this ‘unscaled’ indoor approach to agriculture. Not only does weather (drought, frost) cease to be a factor, so too do insects and the need for pesticides, and without the need for long-distance transport, food stays naturally nutritious.

Satellites too are finding usefulness in farming, such as by combining geospatial data and deep learning. US company Indigo, for example, can predict crop yields via satellite and one day, says Wired, will be able to “look at any cultivated field on the planet and know what crop is growing there, when it was planted, what kind of soil it’s growing in, how well it’s growing, what the protein content is, what the yields will be, and when harvest time will be.”9

From space.

Reducing the environmental impact

Digital technologies can also reduce environmental impact. The same innovations that helped boost yield intensity in the mid-century, such as insecticides and herbicides, are often damaging to the environment. Due to the nature of their current application – en masse, untargeted and sprayed over vast areas – pesticides are not only inefficient and costly, but can pollute natural ecosystems and water sources, and are suspected as the culprit in falling insect and bird populations.10

Precision agriculture on the other hand, such as targeted on-leaf applications of pesticides, or weeds identified by AI and individually lasered, has the potential to reduce the cost to farmers of expensive chemicals, increase yield (by some estimates up to 70% by 2050) and profit.11 As these solutions mature, they are also lowering the costs of the next level of automation.

Farming closer to, or even in, urban centres will also have a positive effect on the environment. Without the need to transform vast areas of land, soil erosion and compaction which contributes to flooding, becomes less of a problem. Not to mention that growing food closer to where it is eaten means less transport and subsequent greenhouse pollution.

Of interest, the same study that predicts a move away from a meat and dairy diet would free up billions of hectares of land, also believes that such a shift would reduce food’s greenhouse gas emissions by 49%, or 6.6 billion metric tons of CO2eq. It’s food for thought.

Robots, and sensors, and drones, oh my!

Despite the challenges faced by farms, technology does offer solutions. While many of the above examples are still being trialled and refined, the need for increased food production and better environmental care remains a certainty, all but ensuring that further development take place.

Farms of the future, be they inside or outside, will need to be adaptable as conditions change and the demand for yield increases. While seemingly the antithesis of digital, rooted as they are in soil and bone, farms have a long history of benefitting from, and driving, technological advancement.

For farmers and consumers alike, this trend must continue.

*The question of who owns this data is another, more complex issue.

**As Brexit unfolds in Britain, and many countries wage war on immigration, the potential loss of migrant labour has farmers asking if their fruit will wither on the vine. While trials have already proven grains like wheat and barley can be planted and harvested solely by automated machinery, more complex crops such as vegetables and fruit will require bespoke solutions (read: costly) and more advanced technology.

***Of course, new problems do crop up. Cost for an unproven technology, for example. ‘Farming as a service’ options, allowing tech to be leased by area or year help reduce farmer outlay and improve willingness to experiment. Additionally, with farms located away from urban, well-connected wi-fi or mobile network areas, simply uploading to the cloud is often not an option. Roaming technology either has to store and upload its payload once it returns to a connected hub, or better connectivity needs to be implemented.