The mavericks of DIY innovation are blazing a trail through the heart of agriculture. Yesterday’s technological hopes are today’s reality on many operations, evidenced by a growing number of farmers involved in automation and open sourcing. If the maxim of “money talks and theory walks” holds true as a measuring stick, then the increase of farmers transferring workbench prototypes to field activity is a plain indicator of success.
Geography once ensured the isolation of DIY innovation, as each farmer tinkered on his own island, but the physical barrier of distance has been spanned by the wireless wonder of a cellular signal. With smartphones in pockets and tablets in cabs, farm inventors from Mississippi to Manitoba are thriving.
“Some guys will continue to pay $500,000-plus for a new tractor and all the extras, and others will buy an older model and dress up it with open source and DIY,” Brian Tischler says. “That’s farming.”
Located roughly two hours east of Edmonton, Alberta, Tischler is driven by the possibilities of open source technology, and far removed from the comforts of armchair innovation. Growing field peas, fava beans, wheat, canola, barley, oats, flax, and sunflowers on 2,500 acres outside the tiny town of Mannville, Tischler is boosting open source access with AgOPenGPS—a free software program aimed at precision mapping and tractor automation which has gained global traction.
The potential for DIY-related savings and efficiency is exploding, partially due to a radical decrease in hardware costs, says producer Brian Tischler, designer of AgOPenGPS. (Photo credit: Brian Tischler)
Tischler began the project in 2016, initially creating a basic application on a Windows tablet that took GPS data and drew a continuous line showing where he had seeded last. “Agriculture is so proprietary and locked down solid. I thought it was time to go open source.”
He placed the entire project on GitHub for free download, and posted it on The Combine Forum. “I said, ‘Here is a link and it does mapping.’” In short time, Tischler added section control and autonomous tractor control to the program. “There are lots of commercial systems that do all of this, but they are expensive. This one is free.”
Tischler doesn’t shy from blunt assessment or criticism: “I want to give back to agriculture because I’ve been very fortunate in so many ways, but I don’t think the future of farming technology is based on open source; it only has a role. Manufacturers continue to make software that runs with ease—the push of a button that does 800 things in the background. Everyone in ag has to make money.”
The potential for DIY-related savings and efficiency is exploding, partially due to a radical decrease in hardware costs, Tischler explains. The pièce de résistance that has fueled tremendous innovation, and is almost ubiquitous among DIY farming advocates, is the Arduino, a simple, open-source microcontroller originally designed in Italy for high school learning. Essentially, an Arduino is a highly durable microcontroller on a board enabling basic understanding of how a computer works. “They took a $3 computer and built software. Now the silly little thing is used on an inestimable amount of projects around the world, including farming.”
Tischler used an Arduino to connect with a driver which powers a tilt meter and an electric motor turning his John Deere 4560’s steering wheel. “It’s $3 for an Arduino, $3 for a tilt meter, and $25 for the motor driver. You now have autosteer. People are absolutely getting bolder with DIY projects by the month. DIY guys have never had this many affordable parts and pieces available, and it’s happening everywhere.”
Agriculture is at technological tipping point, according to Kyler Laird, an Indiana farmer growing 1,700 acres (Lairdscape) of no till corn and soybeans in Jasper County, halfway between Chicago and Indianapolis. At the vanguard of ag automation, Laird has developed a series of DIY robots, starting with a John Deere 420 lawn tractor and continuing up the driverless ladder—Massey Ferguson 2745, Challenger MT765, and John Deere 6330. In 2017, Laird planted his corn fields (535 acres) with a driverless tractor. In 2019, under the banner of his fledgling company, Sabanto, and alongside business partner Craig Rupp, co-founder of 640 Labs, Laird aims to plant 10,000 acres of soybeans from Texas to Canada in a planting demonstration of equipment utilization and robot efficiency.
See here for a collection of Laird’s tractobot videos.
Laird’s “tipping point” analogy rests heavily on the marked decrease in hardware prices over the past decade. “I see more and more DIY. We’ve got access to cheap RTK and GPS, and that hasn’t been the case before, and now it’s a big difference. Anybody can buy a Raspberry Pi for $30 and put together a system. The tech has been there for 30 years, but now it’s off-the-shelf easy items. Almost anyone with a technical bent can do it or quickly learn how to do it.”
“One guy has to make a guide sheet and explain what hardware is needed, where to get it, what software to download, how to build it, and go. Once one person does it, it’s over,” says Kyler Laird. (Photo credit: Anne Bartlett Photography)
Laird is frequently contacted by producers trying to save costs and steer away from subscriptions. “Often, guys don’t want to spend $4,000 on a guidance system and they’re frustrated by the expense of subscriptions. They want something that’s cheap, but works and is functional. Like never before, that is now possible. Then there are guys who aren’t necessarily as concerned about price, but really want more control, and that’s exactly how I personally got involved with automation: I wanted control.”
The open source collaboration of farm technology is set to jump, Laird contends. “You’re going to see more projects like AgOpenGPS. Someone is going to come in and start banging out better hardware for these type of projects, and it’s going to make things very accessible for a lot of people.”
Laird cites the success of Purdue University’s open source ISOBlue, comparing its functionality at a level similar to FieldView. A significant open source issue is hardware access and setup, he explains: “One guy has to make a guide sheet and explain what hardware is needed, where to get it, what software to download, how to build it, and go. Once one person does it, it’s over. For some people, there’s no reason to buy systems of any kind when they can build for much less.”
What did it take to make a driverless tractor in 2015? A batch of free software, some drone parts, a tablet, and one curious farmer to cobble the bits together. Matt Reimer, 31, built automated controls for his John Deere 7930 and uses the driverless vehicle to haul a grain cart during harvest.
“I still think grain harvest is one of easiest areas of ag to automate. I’ve seen where one tractor operates behind another, with the lead tractor having a driver, while the second tractor is automated. That makes sense to me for the future. I don’t see dropping a machine in the field and letting it go entirely solo, but I see automation always needing a person in a lead vehicle to solve problems.”
“We’re at one of those times where tech reaches a point and people figure out how to combine things from multiple sources,” explains Matt Reimer. (Photo credit: Matt Reimer)
A pioneer in automated tractor technology, Reimer, 31, farms in Killarney, Manitoba, and makes heavy use of Arduinos, monitoring tank fluid levels, controlling water pressure, and operating pumps. In addition, Reimer installs sensors and writes the instructions in code to turn on warning lights, and send texts or emails.
“I’ve sometimes felt I’ve worked in a vacuum, but it is always surprising when you find out somebody is working on something similar and you can see how your efforts have helped. We’re at one of those times where tech reaches a point and people figure out how to combine things from multiple sources.”
Currently in the market for a new tractor, internal cab component costs are a heavy source of frustration for Reimer: “It’s $20,000 or more to set up autosteer and a lot of the time it’s just a software unlock code to make the tractor work with other current equipment or your desired level of GPS activity. It’s mind boggling to pay for hardware and then have to pay for an unlock code.”
Steering and implement control via monitor is an area ripe for upset in the ag market, Reimer asserts. “Monitors might cost $7,000 or more, but do nothing more than an iPad. Someone is going to figure out how to stick an iPad in a tractor, open a single app, and let it do everything. In turn, that opens the door to developers all over the place instead of specific dedicated hardware. In the near-term, instead of dedicated monitors, we’re going to have iPads for a tenth of the cost and still be connected to the net.”
Perry Casson, 54, part of an emerging group of growers versed in farming and computer technology, grows small grains—barley, canola and wheat—near the town of Medstead in west-central Saskatchewan.
In 2015, he owned a three-combine fleet, but lacked a single working yield monitor. The gap drove him to build an affordable monitor that uses a display most farmers already own, a smartphone. Casson’s DIY prototype resulted in a commercialized product which debuted in 2018—FarmTRX, an easy-on-the-wallet yield monitor system costing under $2,000.
“This started when I had a measurement problem on my farm and I had the tools to fix it. My tool kit is a bit different than what many consider typical for a farmer in terms of things like hardware design and software development, but the tools to build these kinds of things just get more accessible all the time. A younger, more tech-savvy breed of farmer makes me pretty confident this is just beginning.”
Recently, Casson wanted a test platform for a new RTK GPS receiver he and his team are working on, so in 2018, he began crafting a self-driving mower to cut his airstrip using drone technology and open source software packages. “It’s almost an assembly process and not an engineering job. So much has already been figured out for you.”
Casson is planning to test the mower in spring 2019. “Guys have already done this; I’m just following in their footsteps to build one for myself and wanted a useful device that allowed us to test some things. It’s so cool because technology like this is already good enough to build useful machines for the farm and it’s only going to get better.”
Driving an Allis-Chalmers D17 across his South Dakota farm lot in mid-January, exposed to a heavy wind and snow, agriculture’s version of MacGyver isn’t averse to old equipment; he thrives on it. Jared Schott, 50, grows 2,000 acres of corn, soybeans, sunflowers and wheat (in addition to Limousin cattle and commercial Angus) on land just west of the Missouri River and north of the Grand River at the north-central tip of South Dakota. Equal parts farmer, rancher and tech cowboy, Schott is hard-wired to improve the mechanical components of his operation, and he’s done so with a never-ending DIY stream.
South Dakota producer, far left, is hard-wired to improve the mechanical components of his operation, and he’s done so with a never-ending DIY stream. (Photo credit: Jared Schott)
Whether spending $20 on old backup cameras and monitors to create an equipment surveillance system to monitor booms, hoppers or belts; building an automatic gate entry for pasture access with parts from a remote control vehicle and lawnmower axles; or creating an online database (shopdrusa.com) with a $15 barcode scanner to search his inventory with a smartphone; Schott is equally comfortable behind a keyboard, in a cockpit, on a tractor, or in the saddle.
Schott graduated from college in the middle of a tough farm economy in 1991, and walked into the tech world, unaware his farming absence was merely an interval. He worked for several digital companies, moonlighting as a mechanic in a Harley Davidson store (and getting paid in motorcycle parts). Kodak caught wind of Schott’s ability and sent him to Washington, D.C., where he was contracted to work on software at the Treasury Department, Justice Department, Pentagon, and Quantico. All the while, Schott learned skills to use on his farm 1,500 miles to the west. At night and on weekends, Schott built his own airplane using ribs from Cubs and other old models, and eventually dropped in a drone engine bought from a military testing facility. Built and flight-tested in D.C. just after 9-11, the aircraft currently plays a big role on Schott’s South Dakota operation.
Farmers need a “go-to” forum for inventions and innovation resources, Schott contends. He wants to create a website (farminvent.com) as a central repository of agriculturally-related tech ideas. “I want to build a database of farmer inventors, with an app as well, so guys looking for a particular type of farming invention can find it easily: Machinery, 3D printing, robotics and everything else.”
Each winter, Schott searches for like-minded innovators and DIY opportunity. “We’re looking for ways to build profit per acre. Farmers are the best resource for other farmers.”
In 2018, he bought a FarmTRX yield monitoring system for a 1680 Case IH combine, and hopes the device is part of a coming wave of similar cost-effective technology. Schott adheres to used equipment, purchases old combines at $5,000 or less, makes all necessary upgrades, and drives them until they die. “It’s extremely expensive for the average farmer to buy in to new technology. I see more guys getting bold with their shop solutions and there is some impressive stuff out there. I think some of these guys didn’t have the opportunity to farm out of college, and were trained in electrical engineering and similar fields. Now they’re coming back with expertise and connecting it to technology already available.”
Jim Poyzer grows corn and soybeans outside of Boone, Iowa, and is crafting back yard digital solutions with a keyboard and hammer in hand. In 2012, ahead of the DIY curve, he spent $300 and used a microprocessor to build his own planter monitor for a 1969 John Deere 7000 planter with adjustable corn meters. In the spring of 2015, he began experimenting with variable rate technology to compensate for sandy areas of lower production in his fields, and wrote a GPS-responsive program to plant according to prescription.
This spring, Poyzer is building a mesh network (eight monitors) of temperature sensors and moisture probes that will report data to his smartphone and website (outfarming.com). Each monitor, including a case made by Poyzer on a 3D printer, costs less than $50. “For such a low cost, I will be able to know the condition of my fields before planting. As I continue to learn about mesh networks and sensors, I plan to post and basically give away the code I’m using and how to hook it up.”
Iowa grower Jim Poyzer is building a mesh network (eight monitors) of temperature sensors and moisture probes that will report data to his smartphone and website. (Photo credit: Jim Poyzer)
Initially, Poyzer, 68, delved into 3D printing to connect electronic flow meters together. “I have liquid fertilizer on my planter, both in-furrow and 2x2x2. I’m migrating it all to electronic monitoring. The flow meters hook on my planter and talk to a tablet with an app showing the flow so I can monitor exactly what is happening.”
Currently, Poyzer is fine-tuning a RTK system with a fixed base and a rover. Beside his workbench, he has a box of parts for an RTK correction system that uses a cellphone to receive data, instead of a base station, getting a free correction service from a local airport. “It’s an extremely accurate alternative to a base station. The comparable price off the shelf from a major player was $10,000-$15,000 plus a subscription cost. My parts? Somewhere in the range of $350.”
Poyzer’s 3D printer (Creality Ender 3) cost $200, and the farm applications are “simply amazing,” he describes. “With 3D CAD software I can make almost any shape and even put in threaded holes. I've watched YouTube videos on how to make things out of plastic, place them in a mold, and then remold them in aluminum. I can print objects as large as 9”x9”x10”. This is no toy; this is a tool for me.”
Learn Code, My Child?
Considering the pace of change, what skills might tomorrow’s farmers need? Start with the basics, Laird advises. “Just begin a kid with the basics of code and programming. As things develop, I think you’ll be able to order hardware and it’ll be just a matter of putting it together; you won’t have to know everything. A basic understanding of code and programming is a valuable tool so you can jump in and change or fix the pieces when needed.”
Schott sees an inevitable move toward code instruction: “I can’t imagine any engineering side of any college not teaching you to buy a $20 processor off Amazon and being able to write some code. Kids in kindergarten are going to be able to do this at a basic level. To be able to write code and write your own custom application is going to be very advantageous.”
Reimer believes a mixed skillset will be necessary, but remains uncertain as to what pieces will be most vital. “It’s tricky predicting what a farm kid today needs for tomorrow. Knowledge of programming and mechanical ability is going to factor in, but I’m not sure at what level.”
The Big Tomorrow
How does the near future shape up for DIY, open sourcing, automation and innovation? Tischler sees huge growth potential in weed control. “One big area for future application is robotic crop tending; making your own computer vision to remove everything but the crop. I don’t say this because chemicals are bad, but because we are running out of chemical options. Herbicide resistance may be an area that benefits from automation that people aren’t thinking about.”
Another of Tischler’s concerns: variable rate prescriptions. “The software for VR is so locked up. We send data to the cloud and then a company comes up with magical prescription based on who knows what. Instead, I want automation to get the farmer directly involved in deciding what rates and where to apply.”
Agriculture is entering a period of unprecedented opportunity for DIY farmers or modest engineering firms to develop technology, according to Casson. “This is a golden age. All kinds of robotics, sensing, and communication technologies are just so accessible now. Literally, there are parts out there that now cost $100 that cost $10,000 just a few years back. It’s going to be a fun next 25 years and I want to be around for it.”
A large increase in intensive farm management is coming, Reimer adds. “I think we’ll still go out and plant a single crop in a section, but we’re going to seed at different rates and apply chemicals at different rates. I think smaller, automated machines will be involved, but we are quite a ways away from that whole package. DIY is on the uptick, but even if a given innovation works on a farm, it’s still very tough to convince other guys. However, if something genuinely helps with profit, then other guys eventually will follow.”
“Put yourself in the position of using a horse when then stationary engine came out,” Reimer concludes. “In hindsight, it’s always obvious where technology is going, but it’s not so easy to predict in the moment.”