How smart agricultural technology is driving sustainable farming

A burgeoning example is the development of vertical farming. Typically located in warehouses close to city centers, the environment in these farms is closely controlled and monitored by strategically placed sensors that collect data points on crops. The information is then processed by AI and machine learning algorithms to cultivate optimal growing conditions.

Due to their closely monitored and controlled environment, vertical farms typically use fewer resources. For example, US vertical farm company Plenty says its technology can yield as much as 350 times more produce in a given area as conventional farms using only 1% of the water.⁶ This is notable considering the World Bank reports that farming accounts for 70% of the world’s water withdrawal via irrigation.⁷ As climate change disruption increases, so will the cost of and competition for water.

To use less of this precious natural resource but still yield first-class crops, the company deploys enhanced sensor systems that monitor the health of each individual plant, ensuring each receives the exact amount of water it needs, mitigating waste. It also uses LED lights in the ideal spectrums and amounts to create photosynthesis. The photosynthetic wavelengths are synced with the crop’s growth to minimize energy usage and optimize yields.

Other companies developing vertical farms, such as Iron Ox, take a “robotics-first” approach, using robots with cloud-based “brains” to automatically manage the crop-growing system.⁸ The US imports around 32% to 55% of the fresh fruits and vegetables its population consumes,⁹ and vertical farming proponents say it can help reduce this reliance on imports, as well as cut carbon emissions by shrinking transit times. Additionally, because the crops grow upwards, often on horizontally stacked shelves, they require much less space per crop compared to conventional farming operations.

The use of AI in agriculture is set to rise. A report by Million Insights estimates that the global artificial intelligence in agriculture market will reach $2.9 billion by 2025, with a projected compound annual growth rate of 25.4%.¹⁰

This extends to IoT technologies and robotics. A “connected farm” with sensors and dedicated 4G or 5G networking can deploy wireless (and often solar-powered) robots and devices to improve the efficiency of crop management. The higher speed and lower latency of 5G enables near real-time data transfer, helping to ensure the safe use of robotics in an agricultural setting. Two examples on the market today are Ecorobotix¹¹ and Naïo,¹² both autonomous weeding robots that use cameras, AI and machine learning to identify and treat weeds in the field. Where possible, shifting jobs like weeding to an autonomous smart robot can save time and reduce laborious work.

Similarly, new devices that use GPS technology to track assets like cattle across expansive farmland are helping farmers more efficiently monitor and reduce the loss of livestock. These devices, in part, delegate health and location monitoring of cattle to agritech devices reducing the burden on farmers.

For example, ear tags, such as those developed by Moocall, geolocate a cow as well as monitor eating, drinking and other behavior to assess their well-being.¹³ This technology can save farmers considerable money by flagging health issues early. In the UK, it is estimated that livestock farmers lose £270 million each year due to parasitic worm infections alone.¹⁴ In addition, the tech can check when a female cow is in heat and a pregnant one is ready to give birth. GPS tracking devices can also be used for technology fleet management so farmers can identify where vehicles are at all times.

Cellular-connected drones equipped with cameras can also be used to quickly inspect a large expanse of land to closely monitor crop fields and cattle. The footage can later be reviewed using cloud-based video analytics and AI to identify anomalies. Approximately 20% to 40% of the produce from cultivated crops die each year globally from diseases and pests;¹⁷ however, specially equipped drones can help spot issues quickly as they arise and before they manifest.

Technology examples in this field include Israeli-startup Taranis,¹⁸ which images whole fields with light sport aircraft and drones. Using image-processing algorithms to stitch together the pictures, it produces a unified field photograph down to leaf level to identify zones in need of attention. Combining this with granular field-level weather forecasts, it recommends the most effective time to apply treatments or highlights the best planting window. This helps reduce the waste of costly materials due to unforeseen weather conditions.

EarthSense has produced a ground robot that generates a detailed portrait of crops, including height, condition and leaf-area index, in the field.¹⁹ Called the TerraSentia, it uses a mix of sensors—such as visual cameras, light detection and ranging (LIDAR) tools and GPS devices—to autonomously collect data on plant health, physiology and stress response. Its cloud-based platform enables agronomists to teach the robot to automatically measure a range of key traits.

Such detailed information, which farmers in the past would have collected by hand, helps agronomist select the best plants to use to create hybrids or for producing more resilient and sustainable crops in the future. The data can also be fed into algorithms to understand how future crops may behave and be managed more sustainably.