if I had to replace my tractor today I honestly have no idea which I would pick, but I think all of them come with irreplaceable software components. Are there any tractors available that have fully reviewable / replaceable control software?
There was a project quite a few years back called “open source ecology” where the goal was to develop everything needed to build a functioning farm from scratch. One of the parts involved a tractor of sorts.
However, I think the project lost steam years ago and has since shifted focus.
EDIT: Found it
It seems that the tractor actually went somewhere: https://opensourceecology.dozuki.com/c/LifeTrac
i am bad at reading when i’m tired, apparently. i scanne. through this thread looking for open source ecology before making my comments. my apologies for the duplicate link
Woah that is a very cool, ambitious project. Thanks.
https://opensourceecology.dozuki.com/c/LifeTrac
One of the iterations of this was designed by some of my classmates at my Alma Mater about a decade ago (a little more).
Just learned about this project today and it’s more or less exactly what I’m looking for. If I can build something that I can steer and pull implements with, and that I can then mount my own control box onto… One of my long term goals is building some farm bot orchestration software. For example, if a tractor’s battery gets low enough, I’d want it to automatically drop its implement, go charge / battery swap, and either it picks up where it left off or, if another tractor bot with a full battery is available, the task gets handed off to the new bot.
When I was more on the ag side of research, I know a former pilot/engineer who had decided to ruin his sons life by helping him on his farm.
I kid. The guy was actually a fucking gangster. He had built/ modified a customer variable rate drill/ spray rig. The used it in combination with a hyperspectral sensor they had in a hole drilled into the up auger of the harvester. They worked with us to develop a hyperspectral model that used middle infrared light to predict protein.
So basically, you run your harvest, and when you are doing this you are tracking both yield and protein content (with the sensor in the up auger). Then, when it comes time to plant, you’ve got a map of exactly how much nitrogen was used the previous season. Using this you now have the option to create an optimal yield map, or you can target a specific protein content for the grain.
Anyways, no I don’t. But I’ve seen some pretty cool diy shit.
I barely understood any of that but it sounds really cool and I’m glad people are doing stuff like that. Assuming it works of course.
Turn your tractor into a data collection device.
I figured that out. I simply am not involved in agriculture.
That is very cool. I’d like to get to the point of designing and building custom rigs like that. I am currently in the “can sometimes fix it as long as there’s a manual” phase of expertise so I have a ways to go.
What type of farming do you do?
Hay, but I’m looking to branch out. I feel like it’s a good time to learn how to grow people food for some reason.
I have a '90s Kubota with zero software and just wiring. Downside is, even here in Japan, parts for them are getting really hard to find.
Yeah parts are another huge problem. It would be very cool to have a tractor built entirely out of commodity hardware, or hardware that can be fabricated locally e.g. an on-premises shop or a community maker space.
These people are trying to reverse engineer Deere tractors, but there hasn’t been an update in a few years. Maybe Chinese or Indian tractors? The Indian government has a track record of telling ag companies to stick punitive IP restrictions up their ass, but I’m not sure if any right-to-repair mandates would extend to foreign subsidiaries like Mahindra USA.
If I had the hardware to spare (and the time [and the expertise]) I would love to try and hack a tractor and build some free software for it. Eventually I’d like to be able to write simple execution jobs for my tractor (e.g. cut this field, bale that field, etc.) and send it out into the field on its own. I’d also like to avoid paying a subscription fee for the pleasure, which seems like where the manufacturers are leaning at the moment.
Yeah I think the future of what you describe is Chinese. They are already making autonomous electric tractors, but its early. Tractors are kind of uniquely suited to carry big heavy batteries and never get too far from the charging station. There are smaller autonomous vehicles that can do simple tasks like mow, and of course all the good drones are Chinese. I don’t know if they’ll make it open source, but they may not be openly hostile to users like western corporations. Deere and company won’t just have a fee, all the telemetry, controls, everything will be tied into their own proprietary cloud shit, mandatory reliance on the dealer network, coupled with warranty service, etc.
Hopefully it’s possible that even if a Chinese vehicle doesn’t have FOSS software it’s compatible with FOSS route planning software or diagnostics, or whatever. Maybe it’s not FOSS out of the box but if you flash it, it won’t brick itself because your binary isn’t signed with their private key, or some shit. The lack of dealer would suck and of course, the government could just ban Chinese ag equipment when the check from Deere clears.
Yeah China does seem to be out in front at the moment. I think farming is going to change very rapidly over the next while as renewable energy infrastructure builds up and we get better at growing things around or beneath solar panels / wind turbines. I think soon we will have energy independent farms that are increasingly capable of hands off operation, generating higher margins for whoever owns the land and allowing them to undercut farmers using older equipment with higher labor costs, which will accelerate the growth of megafarms. And I think for small farmers trying to stay competitive without signing predatory agreements, we’re going to need open source robotics and agricultural management software to keep up with the proprietary systems, or at least close. We’ll need to be able to build and maintain our own equipment, in an environment where the best equipment designs are highly specialized for their particular region and crop, and might change (relatively) rapidly as new optimizations are discovered.
John Deere, but only if you bought the Chinese Export version where China sad No Thanks to software lockout. But in all seriousness I have not researched other brands
