The focus today is on how artificial intelligence is being applied to agriculture. This includes developing computer vision and AI platforms to support researchers, breeders and farmers. USDA researcher Steven Mirsky says the goal is to integrate AI technology into a variety of hardware systems used in the field.
“A simple little robot that can drive through a field, or this electrical unicycle we call the phenocycle that we built, or whether it’s on a large-scale tractor with a spray rig.”
But for the code writers and programmers out there, this is the part that should generate your interest:
“We have a modular platform that has a single code base that can serve on a wide range of platforms for collecting imagery and other data, and then building intelligence around that imagery to be able to select in a breeding program, to be able to help researchers evaluate plant performance and for farmers to do variable-rate inputs on their farms.”
So, what exactly does this all mean? Well, with the camera and computer tech, the word is consistency.
“It’s the same camera system that can be used for a wide range of applications. And because it’s a closed ecosystem, as far as the technology, it can be mounted on any platform,”
Like that little robot Mirsky was talking about earlier. Not so little: about three feet wide, nearly five feet tall, skeletal build, and sensors covering it all over. So, what does this outfit pick up in terms of data and information a farmer could use for their crops?
“You can be mapping the plants, mapping their performance—like how big are they? Are they under stress? What kind of stress are they experiencing—water and drought stress? Is it nutrient stress? Or you could be quantifying, have you lost plants out there, like you have less plants just because something died? And then you could also be using that to then deliver nutrients or deliver other types of management practices.”
These innovations also focus on affordability and accessibility.
“These are very low-cost technologies. We build all around low-cost solutions. So, our camera systems are only like $300 to $400. All of these platforms are very accessible, and we’re already deploying these with commercial custom applicators at scale across tens of thousands of acres. More and more farmers and researchers are already deploying this. We’ve got, I think, about 27 states across the country that are using this technology today.”