Four technologies disrupting the Agtech space and how to increase digital adoption

Augmented intelligence and machine learning

It should be no surprise that this is the number one technology. As the power of the computing grows and the cost goes down, it becomes more affordable for growers to adapt these technologies to their farming practices. I use the term "Augmented Intelligence" because these are tools that still need humans to make decisions. They can remove tedious leg work, inaccuracy, and human error.

For example, computer vision and object recognition can enable you to use your phone to quickly and accurately count kernels on an ear of corn in order to predict yield. It can use satellite imagery and historical soil survey data to tell you where you should go in the field to sample your corn but only you can decide what to do with that information.

5G cellular

Not only is 5G significantly faster than 4G, with an expected average speed of 10 - 20 times faster but it can also support 250 times more devices in the same area. This will allow more smart sensors to run locally over the same area and feed in more data much faster. Connectivity is always the top complaint and concern when it comes to bringing technology to the field. We can make devices that can withstand moisture, heat and dirt but we cannot guarantee you can get a signal. The 5G standard enables devices with enough power to create a mesh to transmit data from device to device until it can reach a base station. Combining this mesh with low power IoT devices will allow growers to cover areas that are currently unavailable to them. For those consumers who can access it, 5G will be a transformative change in the amount and quality of data you will be able to access in order to make informed decisions.

Edge computing

While 5G cellular is on the way, it will be a while before all of the US has coverage. For crops being grown in South America where there is limited connectivity, Edge computing will be critical to successful agriculture. Furthermore, barns and other buildings will still interfere with 5G signals therefore the need for Edge computing will not go away. Whether it is embedded in a sensor or in your phone, the ability for you to run machine learning in the field without relying on the cloud, will continue to drive value through making more informed decisions when you need to make them. TensorFlow Lite core runtime can fit in as little as 16 KB of memory! You can deploy learning models that can detect disease, crop stress, and other problems that real time without connecting to the cloud.

Digital supply chain

The recent COVID pandemic and the continuing impacts of climate change have shown the need for improving our supply chains. Before now, we had been optimizing for speed and cost, looking for ways to get the product to market as fast as possible with the least cost. Now, we will begin to see the search for resiliency and reduced climate impact come into the picture. We are already seeing smaller millers such as King Arthur and Bob's Red Mill thriving during COVID as more people are looking locally for supplies. Larger corporations will need to not only diversify their supply by region but also be able to track, predict and manage these increasingly complex pipelines. There is a need for an independently audited platform that allows partners and competitors to track the local and global supply of goods. New hyperledger technologies such as Blockchain, Hyperledger or Ethereum could provide the backbone for these systems.

All of the best technology in the world is useless if no one is using it. While there is a lot of technology adoption inside of agriculture, there are barriers to getting people to use it. To paraphrase the Simpsons:

“You know, a town with money an agtech company with new technology is a little like the mule with a spinning wheel. No one knows how he got it, and danged if he knows how to use it!”
―Lyle Lanley

Trust

This is the most basic problem to solve. The growers have one chance a year to deliver a crop and they need to know that you are equally invested in their success and not just trying to sell them the latest idea from the R&D lab. Before they share data with you, they have to understand how you are planning on using their data, who you are sharing it with and how it is protected.

Solve problems, don't sell technology

No one needs AI or machine learning. They need the solutions those technologies can deliver. Spend time with your customers. Understand what is most important to them before you tell them you have the magic wand that is going to make it all better.

Collaborate

Work together to test out a new technology. Co-invest time and money into validating the value of the new technology to the grower.

Thanks to my co-panelists Mark Moran from John Deere and John Rahiya from AGCO. I have stolen their best ideas and made them sound like my own. If you want to remember what life was like before COVID, you can watch the panel discussion from March 4th, 2020 here. No masks! No distancing! Handshaking! (No actual handshaking in the video but it did happen. Trust me.)