Turning to autonomy
Established in 1908, Granstrom Farms was originally a conventional row crop and cow- calf operation. However, the Granstroms have always tended to swim upstream. The Nebraska family’s latest move is a shift to raising only organically certified crops and experimenting with autonomous machinery.
- READ MORE: How to determine field profitability
“One of my and my dad’s character flaws is the fact that we don’t like to do what everyone else is doing,” says J.J. Granstrom, who farms with his father, Johnny. “We like the uniqueness of growing crops organically.”
Rewind to 1997, when the farm converted from growing corn and cattle into a commercial hay operation. By 2002, the father and son were farming about 5,500 acres of alfalfa with customers in 38 states. In 2008, the Granstroms were asked if they would be interested in growing blue corn.
“We planted our first field of organic blue corn that year, and it went great,” Granstrom says.
Because they liked the direction they were headed, the pair kept decreasing their hay acres and adding organic acres. Today, corn, soybeans, barley, and wheat are grown on 12,000 irrigated organic acres. They also own grain storage for all their bushels and recently built a rail facility with access to both Union Pacific and BNSF railroads.
“We also have ownership in an organic grain merchandising business,” he says. “We have changed our attitude about what crops we put in the ground and are going after the contracts and the end users.”
As the farm has transformed, so too has its need for labor and machinery.
“The effects of the labor shortage in rural America are snowballing. There are very few farm kids anymore, and the small towns are drying up,” says J.J. Granstrom,
Couple that with escalating machinery costs, and it didn’t take long for the Granstroms to realize they had to consider once again trying something different.
“Because there are less farmers covering more acres, it makes sense for machinery companies to keep increasing the size and speed of equipment. However, larger equipment weighs considerably more, so compaction now becomes a bigger issue,” he says. “We ‘fix’ that by spending more on tracks or better-engineered flotation tires, which adds to the cost.
“We’re creating more problems and not lowering our machinery costs per acre as much as we think we are,” Granstrom continues. “We watch our machinery costs closely, and it can get obscene with the number of acres we cover.”
Sabanto's autonomous tractors have done every field operation except harvesting in eight states across the Midwest and Texas.
By offering an army of small autonomous machines, Sabanto is hitting the mark on what Granstrom believes their operation needs. Launched in 2019, the Chicago-based start-up has a fleet of 60 to 90 hp. autonomous tractors for hire. It has also begun retrofitting existing tractors.
“We set our sights on tackling labor shortages and resetting the capital expenses for ag machinery,” says Craig Rupp, founder and CEO of Sabanto.
“We see a future of smaller, smarter, less-expensive, and more sustainable swarms of autonomous equipment substituting horsepower and weight for time.”
The autonomous tractors have done every field operation except harvesting for farmers — large and small — in eight states across the Midwest as well as Texas. (See “The Missing Piece: Automating Harvest.”)
Using a farming as a service (FaaS) model, Sabanto autonomously tilled, weeded, and planted 1,000 acres for the Granstroms in 2021. The FaaS route allows farmers to experiment with a new technology without the capital outlay. As they become more comfortable with a service, they can apply it to more acres.
“The job was done at a reasonable speed, and we truly got a maximum return on investment out of our machinery dollars,” Granstrom says.
Rupp says Sabanto’s fee is competitive with average planting rates reported in the 2020 Iowa Farm Custom Rate Survey ($20.40 to $23.40 per acre.)
|The Missing Piece: Automating Harvest|
While tillage, planting, and spraying can be done with a skeleton crew, J.J. Granstrom says the one piece that has yet to be solved is harvest.
“Until we figure out an autonomous harvest solution, we still need people to pilot harvest equipment,” says Granstrom, a fourth-generation Nebraska farmer. “The self-setting combine technology is there. Now we need someone to bring us a three- or four-row combine with a grain tank big enough to make it to the end of the field and dump in mother bins or semipermanent structures that trucks could load from.”
Reaching a Wider Audience
As Sabanto works to become a full-scale service provider, Rupp says the company’s progress has been significant.
“We began with a mobile planting unit that could cover about 160 acres in 24 hours continuously. Where we were then compared to where we are now is night and day. We have made a lot of progress in control and monitoring, and have gone 48 hours nonstop running some of these systems.
- READ MORE: Irrigation: Autonomous pivots
We are completely cloud-based with a web interface,” Rupp says. “We also have farmers deploying and monitoring our system themselves.”
Within agriculture, the biggest challenge is variability.
A farmer’s field is a dynamic thing, with several operations being performed. Cameras and sensors must understand all the dynamics and then translate that information, which is no small feat.
“What the technology is trying to accomplish must intersect with the actual conditions in the field, which are going to change throughout the growing cycle,” says Phil Sawarynski, managing director and co-lead for Trimble Ventures, a venture capital fund. “That requires a lot of data and computer learning, but we are getting there.”
In August, Trimble Ventures and six other investors invested $17 million in Sabanto. Trimble’s fund was created to bridge a gap be- tween research and development and acquisitions.
With 40 years of experience in the ag space, Trimble’s strategy is to be an invited guest, working with start-ups that want Trimble to be part of their journey because of the value the company adds. Categories that pique interest at Trimble include predictive analytics, connectivity, autonomy, automation, and labor.
“What we do and what Sabanto does are very complementary and can help accelerate the delivery of this technology and these solutions to farmers,” Sawarynski says.
Rupp says the company will use the money raised to commercialize the concept, taking it to a broader audience nationwide, and he predicts it is going to become a mainstay in agriculture in the next couple of years.
“Autonomy is a continuation of what we originally set out to do,” says Finlay Wood, general manager, off- road autonomy at Trimble.
“The solutions we will be providing going forward are there to continue to make the farmer’s job easier, automating more tasks. Over time, I think we will see fully autonomous solutions become more prevalent. We really see this as a journey, ensuring that at each step of the journey we are making the farmer’s life easier.”
|Battling Weed Pressure|
A big challenge with organic farming is keeping weeds at bay. Timing and planning are critical.
Ten years ago, Granstrom Farms fought wild sunflowers, velvetleaf, and a few other weeds.
“We definitely had weed pressure, but if our timing was close to right, we could survive with minimal yield drag,” says J.J. Granstrom, who farms with his father, Johnny, in Nebraska. “We are now dealing with this satanic Palmer amaranth family of weeds. There are about four strains that have completely flipped us on our heads. They are hideous!”
As they combat weed pressure, the Granstroms have been forced to rethink their crop rotations, strategies, and timing of planting.
“Sadly, I don’t feel like we’re winning, but we have definitely changed things,” he says. “A fully autonomous machine can give us the ability to turn a tractor loose in a field and cultivate, rotary hoe, tine weed, or weed zap as many times as possible without having a body on the seat.
“If we’re forced into paying $15 to $20 per acre each time for this new 2.0 weed control agenda, then we will success- fully till any potential profit out of organics.
We’re starting to think like organic vegetable farmers, and I’m struggling with being able to connect large-scale row crop organics with a vegetable mentality. Sabanto can help fix that.”
The Farmer’s Shadow
As autonomy continues to evolve, Granstrom can’t help but wonder if farmers will become irrelevant.
“What will anyone need me for if the equipment can get the work order and execute the plan?” he asks. “I don’t lay awake at night worrying about it because I see this transition taking longer than I plan to live, but I feel it’s the general direction ag is heading.”
- READ MORE: Robots on the rise
Gabe Sibley says autonomous technology will never replace the people making the important decisions around how growing takes place.
“The most important thing in the field is the farmer’s shadow,” says Sibley, founder and CEO, Verdant Robotics. “These technologies are simply tools that allow him to do a job more effectively and efficiently.”
Launched in 2019, Verdant Robotics spent the first six months of its existence listening to farmers so it could understand the problems they faced before offering a solution. Much of the company’s core background is in perception technology — how you teach machines to understand the world around them.
“One of the most important lessons we learned is that the solution must drive value; it must perform an action they pay for better than how it’s currently being done,” Sibley says. “What we want to do is collaboratively play ‘what if’ to find value. When we are shoulder to shoulder in the field with a farmer, we can unlock that value together.”
The time spent talking with farmers also revealed that they don’t care about robots that are only capable of automatically going from point A to B, because it doesn’t contribute that much to the margin.
“Often the person that drives the tractor is a very valuable person, if not the farmer himself,” he says.
“We have a strategy called autonomy last. What that means is autonomous navigation last because it’s not a high-value contribution to what a farmer wants. What really mattered more was the jobs being done behind the tractor, so we put all the brains in the implement.”
It also requires complex technology. “The computation, machine learning, and artificial intelligence necessary to create these machines takes serious horsepower and algorithms that didn’t exist 20 years ago, but they do now. It’s why we’re starting to see new robotic tools for farming unfold today,” Sibley says.
Initially focused on specialty crops, the California-based company is being drawn into designing solutions for row crops. Verdant’s machines have logged thousands of on- farm hours supporting commercial food crops, including carrots, onions, and leafy greens. It is currently working with a small number of growers and companies in the Midwest on targeted corn and soybean applications that combine multiple technologies to achieve a higher weed- removal rate and reduce chemical usage by up to 95%.
“I’ve never bought into the argument that growers are not switched on when it comes to being comfortable with technology,” Sibley says. “I don’t think it’s the hesitation to adopt technology that is preventing growers from using robotics. I think it’s ‘Where’s the value?’ If you can show them value from day one, that’s meaningful.”