A couple of days ago, when staring at the greenhouse wondering what in the world could be modified to make the growing process a lot faster, less labour-intensive, and insanely cool as a Capstone project, I thought it would be a possibility to mount a Universal Robot UR10 to a gantry on the ceiling above the bed- allowing it to travel along the ceiling and reach down to either harvest plants or maintain the farmbot. I know this sounds crazy, and many of you don’t know what a UR10 is or have any idea of mounting a robotic arm to a ceiling to be even lazier with growing, but here you have it. How about that, @roryaronson? Either custom printing a tool to remove plants, a drill to tighten or loosen bolts, or the gripper tool for recovering dropped tools, and utilizing the interchangeable toolbay (Just like the farmbot 
) we can make the process a lot faster, while presenting new opportunities for growing. (This is what you get from an overachieving student with a lot of resources in their hands)
Machine maintenance would be a huge one right there, past homing the machine can’t do much to correct itself. You mention tool recovery, and I think that’s a pretty core problem solved. If you can use it to recognize where Farmbot’s home position is and when Farmbot can’t reach it, then have it assist, that would be another problem solved.
From there I think you could easily solve some of the gaps in farmbots automation, just brainstorming here but, if your arm could load the seed trays from a larger library (such as pill bottles of pelletized seeds) you could cut out one more human interaction and get plants growing completely automated. From there farmbot can do most of it, except feed, harvest, then clear the bed.
I’ve been toying with the idea of putting a spray nozzle on a second line or go ahead and convert to a utm and try to make a spray nozzle head on my farmbot, in order to foliar feed, with a container for fertilizer, a small high-pressure 24v water pump, a smaller solenoid valve wired on the same open PIN of the ramps. If you did something like this with a robot arm you could go a step further. You could use the arm to mix different solutions, such as BT, Neam Oil, Fungicide, whatever else you use that’s water soluble, or perhaps load premixed containers Cutting out another couple necessary human interactions. After that harvesting like you already mentioned.
Past all that though, the plants grown and spent, and in the way, I personally feel the best way to attack this to go modular, have the arm be able to move either fresh growbags or similar into position when needed, then remove the old one from play. If you don’t go modular, you’d have to use the bot to remove the plant, I personally would go the no dig route, have the bot cut the plant at the surface, then compost top, I’m not sure how you’d get it to move compost though, maybe have it lift prepackaged sacks and cut them?
After that its all kind of rinse and repeat I would think, From planting to soil renewal.
The arm has a camera that we added that we can calibrate to recognize different objects, so we can program it to do work whenever it sees something to do. The arm is very powerful and can lift a load of more than 60 pounds, so depending on how we set up the bay for maintenance and tools, we can do a lot of crazy automation with this.
I’m personally working on creating a custom tool that can dispense powder fertilizer, and I need to modify a toolbay to have it even work. I think I might be able to program the arm to add and remove soil, I just need to see how- a shovel might be the only way to go.
For plant removal and fruit collection, this might be harder. The arm basically needs to be at a level that it can extend and collect, and the plan for the bed that will be built is around 2 1/2 feet off the ground. (The greenhouse’s effective height is 7 feet, and the farmbot Z axis needs 5 feet of space to be used)
Basically, the effective working length of the arm is 30 inches, and the ceiling rail that runs along the length of the greenhouse is 12 feet tall. (12-2= 10 feet, and the arm needs a gantry to move around, so I’m guessing the arm will still be too short- 3 feet over the bed. This might be replaced if the bed is taller, I can reduce the space and raise it off the ground by another foot if the bed is in the centre. (The greenhouse I have is 24’ x 18’ x 12’ but the effective height is 7’, since it’s built like a barn. Luckily, I can pretty much have the arm gantry mounted 2 feet off the ceiling rail).
I will keep you updated, send you photos, etc.
I’ve been pondering the same issue of the farmbot’s high operating height. Though mine has been in regards to grow lights. I’ve been conceptualizing a sort of lighting boom that could be raised or lowered, but I haven’t figured out what I would use to actually give the Farmbot the ability to do so, maybe a 24 servo or winch or another stepper motor. But my idea is have the machine go home, then lower the z all the way, followed by lowering the lights into place. How are you moving the arm around the X and Y plane? If your building a motion platform, is a z axis not possible? Keep your X and Y plane raised, but have a Z like the Farmbot or maybe MPCNC and just move it to a safe spot when not in use. I mean they run CNC routers on walls im sure a stable enough system is possible.
You’ve lost me on your idea of a ‘lightning boom’. The arm I think can be adjusted by using different size mounting offsets, since the idea to to have only an x axis gantry for it- it will travel along the length, and the arm can reach for different areas. A Z axis for that would be a nightmare, and I think a simple metal scaffold to lower it further is more effective. I also have an idea of a sort of flat extension that the Z axis will unfold from, so in a sense a sort of system that will allow it to lower itself without having the absurd extra height.
This topic was automatically closed after 30 days. New replies are no longer allowed.