September 17, 2024
Q&A With Deere on Technology: Cameras, Autonomy, Algorithms

DTN/PROGRESSIVE FARMER: Talk to us about the new S7 series combine.

AARON WETZEL: The S7 series features sense and act capability that help customers improve their capabilities and optimize their harvesting operations. We have a series of cameras outfitted on the front of our harvesting solution that is looking out about 30 feet in front of the header — identifying the crop that is coming toward the machine and automatically adjusting the speed of the machine and the settings of the machine to optimize performance and productivity.

DTNPF: Explain the importance of camera technology as Deere moves towards autonomous systems.

WETZEL: (They) are foundational to us to ultimately get to fully autonomous production systems. Being able to understand the machine’s environment and surroundings, being able to understand the job that is being done and the quality of the job, we’re going to leverage as much as we can with camera technology to sense and act, to adjust our performance in the field. Cameras are pivotal. With these capabilities, we are seeing up to 20% overall improvement at harvest.

DTNPF: How is the 9RX series tractor wired for autonomy?

WETZEL: The 9RX 830 is autonomy ready. Customers can choose the option to have in the machine from our factory all of the wiring and some of the components needed, so that when our autonomy stack becomes available they’ll be able to plug that in and go to the field in an autonomous manner.

DTNPF: Deere’s See & Spray system is built to get smarter with use over time. Fieldwork essentially trains the algorithm to distinguish between plants you keep and plants you kill. How do you train sense and act systems?

WETZEL: We use our cameras to take images of the field, every single plant. We use edge computing to decipher in real time if that plant needs to stay or be killed. As those machines go through the field, we (also) are cataloging all these images, every time the machine passes. That improves our insights around the quality of the plant at the various growth stages, whether it has disease or not. We continue to improve our insights with our machine learning algorithms that enable us to further enhance and make better, year over year, our See & Spray capabilities. We’re going to take that same sense and act technology and bring it to many of our machines.

Dan Miller can be reached at [email protected]

Follow him on X, formerly known as Twitter, @DMillerPF

link

Leave a Reply

Your email address will not be published. Required fields are marked *