Can AI help make farming more profitable

The world needs food — grains, vegetables, et al — to feed the 10 billion-odd people that we’ll have in the world by 2050.

That, according to the World Economic Forum, means that demand will be 60 percent higher than it is today, while climate change, urbanization, and soil degradation shrink the availability of arable land. Water shortages, pollution, and worsening inequality make the problem even more complex.

The way to overcome them seems to be by using new technologies to maximize output and minimize wastage. New-age technologies not just make life easier for farmers but also help ensure that the world will have enough food to feed its inhabitants in the coming years.

Enterprise software vendors like Microsoft and MicaSense and drone makers such as DJI and PrecisionHawk are working together to create exciting solutions that quickly learn about the vast farmlands and help make crucial decisions about soil moisture and temperature, pesticides, and fertilizer — and maximize production.

These new-age solutions make data collection easy and eliminate the hours and days that farmers previously spent walking or driving the fields to try to detect problems.

At the core of these new “agri-tech” solutions, there’s a drone or unmanned aerial vehicle (UAV) with specialized cameras mounted to capture thermal images of the land. They fly above the farmland on a pre-programmed flight path, record video which AI and algorithms convert into data, and turn into “actionable insights” for the farmer.

Microsoft, identifying that wi-fi and internet connections are spotty in farmlands, is leveraging TV white spaces, a type of internet connectivity similar to Wi-Fi but with a range of a few miles, to send large amounts of data from ground-based sensors, tractors, and cameras to a computer on the farm.

It’s exciting because TV white spaces are unused TV broadcast spectrum, which is plentiful in rural areas where most farms are located.

Although agri-tech companies like The Climate Corporation and Gamaya (identified by Intel) already offer data-driven agricultural insights that take soil type, seed suitability, and local weather patterns into account, there’s a lot more that needs to be done in this field.

There’s scope for a lot of growth, but development needs to pick up now if the target for 2050 needs to be met. The involvement of enterprises such as Microsoft and Intel can really help, provided their interest is sustained.