How 5G
will impact
the future
of farming

The escalating climate crisis is killing the yield of the world's most farmed crops, such as wheat and soybeans. Yield forecasts look dismal, the Environmental Health News writes—and that's an especially worrying proposition given that close to 2 billion experienced food insecurity in 2019, according to the Food and Agriculture Organization of the United Nations. To feed a population that the United Nations estimates will reach 9.7 billion by 2050, the industry will have to use its resources—particularly its water, fertilizer and labor—more judiciously.

Current farming operations—such as tilling, planting, weeding and harvesting—are already executed at scale on industrial farms. But these processes too often run with a one-size-fits-all approach, leading to food waste. US farms alone waste 20 billion pounds of produce every year, according to Foodprint. IoT sensors, coupled with AI and machine learning algorithms, could fine-tune farming practices to optimize water and fertilizer use, harvest crops only when truly ready and squeeze more food out of every acre.

But harnessing the data from thousands of sensors will require faster networks with greater bandwidth. 5G can be the conduit the farm of the future needs.

Where the 5G impact will be felt on the farm is through its at-scale facilitation of technologies such as the IoT, AI, edge computing, drones and autonomous vehicles. Collectively, these technologies can optimize routine farming tasks.

Climate change has dramatically affected water availability and rainfall predictability. Cutting-edge technologies can help industrial farmers overcome water shortages and extended dry spells. IoT sensors embedded in the soil can measure moisture levels, and data captured from drones can help generate heat maps that highlight problem areas. Advanced machine learning algorithms can process this data and distribute water where it's needed most. 5G's speed and throughput facilitate the transmission of these large data sets.

Machine learning algorithms could be trained on vast banks of images to recognize various weeds. Edge computing devices out in the field could then compare what they see on the ground against what they know and remove the weed. (A similar approach could also be used to harvest crops or pick fruit.)

See and Spray, a machine developed by a John Deere acquisition, operates by this model, using targeted weed killers only when necessary. Selectively addressing weeds requires data-intensive analysis at the edge, a process better supported by the high bandwidth and low latency of 5G.

IoT collars could continuously monitor livestock and alert farmers when something doesn't look right. Robotic milkers free farmers of back-breaking early-morning labor, and machine learning algorithms could help ensure that cows are only milked when they're ready. Robots could do the grunt work, taking in data from IoT transponders and relaying alerts to farmers when protocols deviate—such as if a cow doesn't show up for milking on a particular day. These robots process data at the edge and relay comprehensive analysis back to the owners, but they'll need low-latency, high-volume networks—the kind that 5G can deliver—to do it.