Rowbot receives commercialization support from the National Science Foundation

We are delighted to announce that Rowbot was recently selected for a Phase II SBIR award by the National Science Foundation (NSF). This two-year grant will support commercialization of a computer vision and machine learning based method (or AI) for guiding nitrogen (N) fertilizer application rate decisions for corn production.

Why does this matter? For corn farmers, N fertilizer is typically the most expensive input after seed purchase. And, importantly, N availability generally has the largest impact on yield after Mother Nature (weather). Yet, even today in 2024, it is commonplace for agronomists to estimate that farmers loose 20 or 30% or more of their applied N fertilizer. Those losses lower profitability while also creating water and air pollution challenges that undercut a grower’s stewardship.

Our novel approach begins with creating small areas of N stress using our proprietary robots. Some weeks later, sensors onboard our robots are used to evaluate leaf yellowness on lower leaves—a condition no farmer wants to see, but it is okay in this case because these conditions are created on tiny areas across a normal field. The images below show examples of characteristic yellowness (N stress) and how computer vision and machine learning are used to identify individual leaves. Yellowness scoring is carried out using another computer model (and validated by human scoring of many images). Maps of N stress can be created for test plots, which at this stage were narrow strips the entire length of a 10-ac field.

Finally, these observations of N stress are compared with predicted N stress from an N model to understand how well the N model matches actual field conditions. When a good match is made, then the model can be used to predict the final N application rate for the field. We believe that this approach will deliver higher profits, yields, and stewardship for growers across the Corn Belt.

Examples of N stress observed as characteristic yellowness on lower leaves, from no N stress (left) to high N stress (right).

Example of images processed using computer vision and machine learning to identify individual leaves that are subsequently scored for characteristic yellowness, a key indicator of N stress. "Segmented" leaves are shown in different colors (e.g., partial leaves in dark blue and fully-captured leaves in bright green).

Preliminary results from test field in Rosemount, MN. Nitrogen stress values were estimated using the computer vision and machine learning approach described above. There was good agreement with actual N stress imposed by varying application rates of N fertilizer (N-treatment codes ranged from no stress (0) to high stress (7).

USDA-SBIR support for greenhouse gas monitoring tech

The Rowbot team is excited to have been selected as a recipient of a Phase I small business innovation research (SBIR) grant from the USDA’s National Institute of Food and Agriculture. This grant has enabled us to develop a second generation chamber (FluxBot™) for measuring fluxes of nitrous oxide—an extremely potent greenhouse gas—as well as carbon dioxide. We are digesting results from the 2024 growing season and setting our goals for work over the winter and leading up to the 2025 growing season, which should be a busy one for this project.

The two images below show a FluxBot chamber being tested along side a conventional chamber. Many thanks to our USDA-ARS collaborators in St. Paul, and USDA-NIFA for funding.

Our grand plan for this patent-protected technology is that these chambers would eventually be moved around by a ground robot, enabling sampling at a variety of locations across a field. This would greatly increase temporal and spatial sampling, the combination of which is difficult with today’s technology.

Rowbot secures National Science Foundation funding to improve in-season nitrogen solution

The Rowbot team is excited to announce that we have secured a Phase I Small Business Innovation Research award from the National Science Foundation. We have long been certain that small ag robots will shine when they combine work with data. What we mean by this is that there is a “win-win” opportunity to collect data, process it in real time, and then use the analytical result to improve the quality of the work performed on farmer fields.

This SBIR award from NSF is focused on the short-term research required to commercialize the data portion of our patent protected, in-season nitrogen (N) solution. Producers who apply N during the growing season (a best practice) and want to adjust rates based on the specifics of the unfolding growing season (rainfall, growing conditions) can use a commercially-available model—such as Adapt-N—to estimate up-to-date N predictions. We are extending the value of using a model by stressing small plots on “every field, every season” to ensure that model parameters are correctly set, or to determine the best parameter set through simulation. This will eventually be conducted on-the-fly at a nominal incremental cost to the producer who will stand to reap the economic and stewardship benefits of improved N management.

Rowbot co-founder interviewed on "Engineering Your Farm" podcast

I was lucky enough to be interviewed by Brian Dougherty at Iowa State University for the Engineering Your Farm podcast series. Brian had lots of good questions teed up. The discussion is about how fleets of small, yet powerful robots can move the needle on farmer profitability and stewardship. We talked through our current focus, which is on in-season nitrogen management and cover crop seeding. Thanks to Brian for the opportunity!

Latest Newsletter: Strong cover crop growth!

Strong cover crop growth!

We visited two of our NE Iowa customer fields last week and observed good growth of the cover crops we seeded using our fully-autonomous system weeks before corn harvest last Fall.

The image above lines up with the same part of the field included in the drone footage below. Read the full post and watch the video here. If you missed it, take a look at our our Earth Day post that included an image from a customer field in western IL.

Thanks again to our early customers and partners at Carnegie Robotics, the Walton Family Foundation, and Prof. Scotty Wells’ group at the University of Minnesota.

We remain excited about taking our technology to the next level. Do reach out if you would like to be part of writing a chapter in the future of farming where we drive farmer profits and stewardship (aka regenerative ag) while helping to increase productivity to meet global demands.

Rowbot-planted cover crop report: growth is looking good!

Took a drive today to check out two of the fields that were part of our 100 acre fully-autonomous milestone last fall. The photo above shows a nice stand of cereal rye that is the exact spot featured in the video below. The cover crop will be terminated in the next few days to make way for this year’s soybean crop. Soil health benefits will accrue for the farmer, including an increase in soil organic matter and carbon storage.

Here’s a nice shot from the second field from today. Also cereal rye. We had some seeder issues on this field, but this portion turned out very well. Many thanks to Loran Steinlage (@FLOLOfarms) supporting us through our on-the-fly learning last fall.

Rowbot has solutions for reducing the climate impact of the ag sector

Field seeded into cereal rye about 30 days before corn harvest (Fall 2020). Photo taken March 21, 2021 by Grant Wyffels.

A few facts on Earth Day. More than 10% of the energy in the U.S. ag sector goes into making nitrogen (N) fertilizer for corn. Even though most farmers know that in-season N application will limit off-field losses, most N goes on at or before the time of seeding. That’s like going to Vegas: it is a roll of the dice that the N required by the crop in July will be there. That’s quite a gamble given that N is unparalleled in terms of its impact on yield, aside from Mother Nature (i.e., weather). We know how to manage N better. Rowbot has a plan that should reduce off-field losses dramatically, driving up farmer profitability, while maximizing yield potential.

 With all of the excitement about carbon storage on farm fields in this new era of regenerative agriculture, we need to consider that about 90 million acres of U.S. farmland is in corn each year. And, importantly, farmers lack good solutions for seeding cover crops on those corn fields. Seeding by airplane into mature corn is hit or miss. Seeding after corn harvest is tough moving north because there are only a few days before snow. Again, we at Rowbot have a solid solution to lay down seed prior to corn harvest. Here’s a recent photo (credit to Grant Wyffels) from a field we seeded last fall as part of our campaign to seed 100 acres fully autonomously.

 We’re ready. Help us take this world-positive technology to the next level! Happy Earth Day!

Rowbot Seeds 100 Acres of Cover Crops for Customers

Over the past few weeks, we have been hard at work advancing our autonomy capabilities. At about noon on Monday (10/5/2020), we reached our goal of seeding cover crops on 100 acres of customer fields across the Midwest. We started in central Ohio, moved to western Illinois, and then on to NE Iowa. We are grateful for the support of our early customers, our partners at Carnegie Robotics, the Walton Family Foundation, and Prof. Scotty Wells’ group at the University of Minnesota.

Take a look at the view from behind and from 100-ft above via drone. In both cases, the machine was running in fully-autonomous mode with a safety operator monitoring it from the side of the field. There were no “kick outs” in the first video, and just one in the second (at minute 10)—note that the recovery (i.e., re-alignment between rows) was conducted by the remote operator. Over time, the frequency of these “human interventions” will drop off and will be solved by a safety operator monitoring dozens of machines.

Enhancing Soil-Seed Contact while Cover Crop Seeding with Next-Gen Rowbot

Fresh from the field where we have been testing our next-gen machine. We’re excited about the potential for using the new machine’s tracks to increase soil-seed contact, a key factor influencing cover cropping success. Shown in the video is one solution we are testing where we’ve added cleats that should help work the seed into the soil more than the tracks alone. We’re collaborating with Prof. Scotty Wells at the University of Minnesota on this project that’s funded in part by the Walton Family Foundation.

Next-Gen Rowbot Hits the Field!

We are excited to present footage of our latest machine during testing this week on a Minnesota field. This new machine excels at moving around on softer ground (think: being able to get back on a field soon after a big rain). It’s also much more stable than its predecessor, which runs on wheels. These features contribute to greatly improved mobility, which enables more robust autonomy. We’re looking forward to having it (and a few friends) running on Midwest fields come spring 2020. Our small, nimble machines are a great platform to handle a range of tasks on large-scale, row crop fields, while also collecting valuable data to make the machines work smarter.