Making sure your farming equipment uses the latest and greatest machine learning models can be very challenging. Connectivity can be minimal and expensive if you’re using satellites, says Jason Campbell, director of architecture at Wallaroo.
So his company launched Air Gap Edge Deploy to make it easier to deploy and manage edge machine learning models in environments without IP connectivity. Think oil towers, gas pipelines, power transmission lines, and utilities; and independent equipment in smart manufacturing and smart agriculture.
We spoke to Jason Campbell about solving the air gap and how it can benefit farmers.
Why was the Air Gap Edge Deploy function developed?
“There are many reasons why companies (not just farmers) are looking to deploy machine learning across the air gap. For example, equipment may be offline at the edge, such as in oil towers, gas pipelines, or in agriculture, And maybe unite in a rural area far from a cell tower.In addition, Increase in cybercrime It has led some companies to explore the air gap option to keep their systems secure. By isolating their networks from external networks, they are able to prevent the security vulnerabilities that come with having these connections such as data breaches and ransomware attacks that can cost billions in losses.”
How does Air Gap Edge Deploy work?
“You can train the machine learning model anywhere, in the cloud or on-premises. But from there, you will need a remote connection (for example, to the actual farm equipment where you want to deploy the model. We provide more details on how it works). over here. “
Why is it suitable for crop growers?
“Agricultural equipment is producing more data than ever before. Their farming equipment has become essentially “mobile sensor suites with computational capabilities,” John Deere’s chief technology officer said in an interview with The Verge. The combination of sensor data and artificial intelligence allows crop growers to use just the right amount Only from water, fertilizers, pesticides, etc. for each individual plant.In addition, the crop farmers rely more on robots in picking different crops using computer vision etc.
Essentially, what you get data plus AI as a farmer is greater return with lower input costs. But making sure your hardware uses the latest and greatest machine learning models is very difficult. You should also consider the flow of data to the data scientists who trained the model. All of this sensor data can be gigabytes or even terabytes of data per day.
The connection is minimal and can be expensive if you use satellites. So for the equipment owner or service provider, it is more cost effective to use something like this air gap solution to deploy a machine learning model in the equipment as well as to output production data so you can be sure that your models are still accurate and performing well. Learn more about what this inbound and outbound data flow looks like over here.
What are the costs for a farmer using this air gap solution?
“As you see a greater shift to equipment as a service (EaaS), more of these costs will be borne by owners and providers of agricultural equipment who then sell it to farmers as part of larger smart farming services. So the farmer will pay for AI services, which will include advanced machine learning as part of it. “.
What are the financial benefits for farmers?
“The adoption of machine learning in agricultural production is becoming imperative due to the need to increase food production while balancing sustainability. ML is already making an impact, providing insights that help increase productivity, use less water, fertilizers, pesticides, etc. It is paying off. .”
Can you give a working example?
“We already have examples of agricultural equipment that automatically adjusts the water, nutrients and other chemicals used down to the factory level using sensor data combined with artificial intelligence. In terms of our air gap solution, we don’t have examples to share but we do in other industries, Especially manufacturing.In fact, we work with US Space Force about deploying advanced machine learning for their fleet of satellites.