10 ways AI can make farmers’ jobs easier

For those who work the earth, AI can be an invaluable tool for making that work more efficient, sustainable and productive. AI helps farmers use the same tools they have been using for centuries, but with improved efficiency and focus. Here are 10 tips for farmers looking to use the new wave of AI-assisted farming. 

LET ROBOTS DO THE WEEDING

It may look like a Mars rover, but the BoniRob system is much more down to earth: It’s a robotic system that uses AI to analyze the soil, determine what is growing in it, and eliminate it if required. It’s part of a concept called Spot Farming, where each spot in a field is handled individually, rather than treating the entire field as a whole. The whole process is overseen by robots and AI that monitors each individual plant and feeds, waters, and harvests the crop. In the U.S., a company called Carbon Robotics is selling a similar device called the LaserWeeder that uses AI to detect weeds, then zaps them with a laser. 

IDENTIFY PESTS

Not sure if that thing crawling on your crops is good or bad? InsectID from Iowa State University can help: Take a picture, upload it, and InsectID uses AI to identify the insect, provide information on the role of the insect, and give pointers on how to handle it if you want it out of your crops.

LOW-COST SOIL ANALYSIS

">Farmspace Technologies is a Nigerian company that is developing a low-cost soil sensor called ActionLab that uses AI to help small farms understand how to use fertilizers more effectively and which crops to grow.

WORK TOGETHER

Farming is about more than just growing things: It’s about getting those things to customers. The Indian government has created a project with the wonderful name Saagu Baggu that helps chili farmers in Khammam use AI to better understand their soil quality, the quality of their product, and to find better prices. It currently covers 7,000 small farms, but the Telangana state government is now expanding it to cover 20,000. 

PREDICT THE WEATHER 

Weather prediction is the latest area AI is being applied to, with Google applying its DeepMind system to weather forecasting and Nvidia doing the same thing with their FourCastNet system. Both systems use AI to generate short-term forecasts, predicting patterns up to 10 days in advance. AI also has a role in longer-term predictions and the role of climate change, but this does underline one of the issues with AI: It is great at finding patterns in existing data, but it can’t understand how things might change in the future.

ARE YOUR TOMATOES RIPE?

You know when your crop is ready by the way it looks, but you can’t be everywhere at once. So, why not use AI to help? This study shows how a group of AI scientists developed a system that can decide when a tomato is ripe from an image. That might seem trivial, but imagine combining this with an automated system that trundles around your greenhouse and sends you a report that helps you decide when to harvest? This isn’t the only example of this type: Similar work has been done for tea buds and many other crops. 

HARVEST WITH AN AI ROBOT

Fieldwork Robotics is a U.K. company that is building robots that can harvest things you might not expect, like soft fruits. Their Fieldworker 1 robot uses AI and robotics to detect ripe soft fruits, harvest them with a soft robotic picker, and place them in a tray. It’s still an experimental project, but it is being tested by the largest soft fruits producer in Australia. They claim it will allow a single human operator to oversee the harvest on several fields at the same time. 

DO YOUR COWS LOOK AND SOUND HAPPY?

Cows are noisy creatures, and the sounds they make often indicate how they are feeling. So, a group of researchers used open-source AI tools to analyze the noises cows make and determine how they are feeling. The results are, they claim, pretty accurate because cows make low-frequency noises when they are content and higher-frequency noises when they are distressed.

The researchers used Open AI’s Whisper tool, which converts sound into text. By adapting this to determine frequency, the length of the noise and how often it was repeated, they claimed they could produce a real-time measure of the welfare of the cows. CattleEye is a system that uses a security camera and AI to monitor cows as they pass through a sort gate, or by analyzing drone footage of your pasture. The clever bit is that the system uses AI to analyze how each cow is moving and how healthy they look, producing a measure they call the Body Condition Score. If a cow has a poor BCS, they may be being overfed, be lame, or have other issues. 

PUT AS MUCH BACK IN AS YOU TAKE OUT 

Regenerative agriculture focuses on long-term sustainability by prioritizing the health and vitality of the soil. RegenIQ offers a way to measure this process and provide advice on improving its effectiveness. By combining multiple data sources with AI, it helps track and measure progress over time.

GET INVOLVED

The wonderfully named Centre for the Study of Existential Risk at Cambridge University looked into the risks of using AI in farming in 2022, and found that there are plenty, ranging from reliability of AI models to the cost. One of the major risks that they identified was the problem of data transparency and ownership: Where does the data that AI uses come from, and who owns the technology?

The authors call for farmers to share this data openly, but without losing ownership by setting up data cooperatives. One example of this is in use in Ethiopia and India is FarmStack, a project from Digital Green that lets a farmer share their data but to also set limits on who gets it and how they use it. In the AI world, this data is more precious than gold, and farmers the world over need to understand this, and work together to make sure it is used properly, ethically and for the good of all.

https://www.fastcompany.com/91236048/10-ways-ai-can-make-farmers-jobs-easier?partner=rss&utm_source=rss&utm_medium=feed&utm_campaign=rss+fastcompany&utm_content=rss