Comment by MisterTea
2 months ago
> However, the tech exists for a reason and is not inherently bad, the issue is the lock-in, the lack of choice and interoperability.
These low-tech tractors could become a hot bed for open source experimentation. Nothing stopping someone from sticking a tablet on the dash. You could run GPS harvesting optimization software or some webthing locally. Could be cloud or clever DiY farmers could run their farm off a local instance on a small machine using a WiFi AP atop the barn or whatever.
This was my take as well. How many 3rd parties might be able to bring on upgrades/modifications to a "dumb" tractor to make it smart vs only being able to buy a "smart" tractor from one vendor and be forced into it's rules/restrictions/prices
Plenty of options for putting auto steer on a dumb tractor already exist.
Cheap ones too -- aliexpress has them.
But there's more to agtech than driving a tractor around, a lot of what these big integrated systems do (at the high end) is very data driven -- determining where and how to plant, irrigate, fertilize, etc. There's a lot of integration work beyond just making the tractor drive.
29 replies →
The kid? :)
1 reply →
Years ago, there was a TED Talk[0] from the guy that started Open Source Ecology[1]. The TED Talk was really cool, but I haven't really followed what they did. It sounded promising to have open-source technology for use in this space.
[0] https://www.youtube.com/watch?v=S63Cy64p2lQ
[1] https://wiki.opensourceecology.org/wiki/Main_Page
I absolutely love this vision. He's still working towards the goal. It seems that his vision has problems scaling up though. He seems to mostly still drive this himself.
They have no driving electronics, electronic throttle, ECU controlled injection etc, so you are limited, you can't for example easily make it go constant set speed, because the throttle isn't electronic.
It went a bit too far, optimum would be modern enough to have drive by wire but with open ECU and documentation
You can still control a completely mechanical engine to work with set speeds. There are mechanical governors that can do this, or you can get an electronic component that moves the throttle for you. Fixed speed engines with variable load are much older than the transistor.
It is no harder than doing it with an ECU, except that you need to install a servo or speed governor with hand tools, instead of fiddling with ECU code.
It is far easier for 3rd party stuff to target say open bus protocol rather than a servo + speed sensor pair.
1 reply →
It has a governor.. The P pump 12 valves (and many other multi-application diesels) come with either one of two different governors, an automotive one which has a high idle and low idle, but unrestricted fueling in between. This is what you want in a car or truck where you're controlling road speed with your foot. There's also the "industrial" governor that essentially maps lever input linearly to engine RPM, and endeavors to maintain its set RPM independent of load. This is the kind you find in tractors, generators, boats, etc.
These governors are basically mechanical analog computers which use the inertia of flyweights, springs, and some very clever linkages to do their thing.
I know, I used tractor like this. Governor only keeps RPM, not the air-fuel ratio and a bunch of other emission and fuel usage related stuff.
And it's a bit easier to make 3rd party addons when you just have some open bus standard, not "mount that servo on a gas pedal"
1 reply →
There are already open source auto pilot and cruise control implementations for cars. (Not all cars are supported obviously!) so to have this in place for tractors off the road seems very doable.
Edit: specifically thinking of https://comma.ai/
Well open source AutoSteer exists it has a lot of features like rate control built in to it. The system is called AgOpenGPS it’s very popular for retrofitting older equipment with modern technology.
I would be surprised if this doesn't exist already in some nascent form?
This is an area where you would probably need entire ecosystem of systems that are onboard tractors, but also for the various implements you hook up to it to monitor sowing, fertilizing, spraying etc. Including backend systems that you can either self-host or subscribe to some service that doesn't have awful terms.
It shouldn't take immense amount of capital to make some real progress towards something that can make a difference.
My bet would be there will be a niche for these tractors at hobby farms but the reality is outside of niche goods and hobby farms, farming is about scale and the machines that companies like JD sell help a lot. Sure the tech is locked down but at the scale those players are running at it’s baked into the service contract to minimize downtime.
A niche called Mennonites in general and in particular, Amish in communities that are ok with motors, but not with fancy, fragile electronics.
Right those are niches like hobby farms. There is a market but it has a pretty distinct shape. The broader point was a machine like this is not changing farming.
The beauty here is even beyond experimentation the tech will change repeatedly over the life of the equipment, and you can cheaply adapt to that. There is very little advantage to the modern tractors, beyond luxuries and the finish of a self contained package. Farmers rarely ime prioritize either of these
With high end tractors you can have them drive themselves on the rows based on a GPS map that was created when you planted. That's going to be difficult to retrofit.