Agricultural Bot

Hi Folks,

I'm thinking about forming a project for a small agricultural robot for developing countries. The U.N. predicts the world needs 80% more agricultural production by 2050. I'm not sure if RepRap is the right place for this. I'm a bleeding heart Berkeley educated liberal, now back in the midwest in the small agricultural community I grew up in. The big buzz word here is precision agriculture, and small drones are starting to make their mark. I think small agricultural robots are next. There is no need to have a $50,000 cab on a $350,000 tractor in this day and age. My father grew up using a two row corn planter, I'm thinking along similar lines. Of course it would be open source and there's not a whole lot of other engineers or technical types near where I live so I'd like to open it up online.

I worked as an Aerospace and Mechanical Engineer on various Mars lander missions for 20 years, so right now I'm thinking on the lines of a suspension system similar to the Mars Curiosity Rover (Rocker Bogie) Lots of open source ideas already exist for sensors and actuators. It would be great if it were augmented by solar panels, but Lithium Polymer batteries are pretty good now and improving fast, and solar just doesn't generate enough power according to some back of the envelope calculations..

Just thought I would throw it out there to see if it would generate any interest

---

Mechanical Engineer by day. Maker by night.

Full respect about your work and projects!
Why not some sort of a vehicle with tracks? I think the hardest part in this project will be the software and the sensors for the robot. How it will "find" the agricultural plants?
Best regards,
Bobby

Hi BagoMaker,

very interesting project!

Have you thought about using some off the shelf platform like a cheap quad / ATV?

I think one of the biggest challenges will be on-site repair and spare part supply in 3rd world countries.

Fell free to contact me if you like to discuss the project.

Best regards,

Christopher

I like the idea too, but it will take many robots to replace a highend tractor. Furthermore the robots should be made from biologic materials as far as possible. With solar energy, they wont move fast, a small, flexible transportation system for the harvested materials etc. is necessary.

Btw, wouldnt it be better to learn more about animals, and "use" e.g. ants to harvest our grain and minerals? If we could communicate, and raise understanding of animals, train them, and be friends, we could do lot more for each other.

Edited 1 time(s). Last edit at 06/06/2014 07:04AM by Kolbi.

I appreciate your work project.May the almighty and your government be on your side.

The bot as I see it would have to be modular and since you are talking 3rd world country run on a variety of fuels/power sources.I am thinking something about the size of a small garden tractor. it would have to:
1. Know where it is and what to do. This could be done with a smart phone GPS + solar powered local wifi transponders. You calibrate it then spot check periodically.

2. Determine the condition of the soil to see what crops could grow there. Smart phone again soil sample is loaded into a hopper then the camera takes photos with a microscopic lens which are sent back to a computer that compares them to a soil database. Other sensors to determine PH and moisture content could be added in.

3. Work the soil. this would be a separate module akin to a rotor tiller add on with a fertilizer hopper, Then another module with a seeder to furrow, plant and disk.

4.Weed patrol. This might be done several ways, the simplest is that the bot just digs up anything not in the row with the tiller module.This leaves any weeds growing in the row itself for a human to come by and weed. A more complex solution would be leaf/color recognition software and a weeding tool.

5.Water/sprayer. Since we are talking 3rd world here there would have to be a pump and reel of hose/water tank if water was available on the surface the water would be drawn through the hose and sprayed behind the bot as the bot wound up the hose. A tank would carry water spraying it behind the bot. The trick here is to get the soil nice and wet but to keep the bot out of the mud as much as possible.

6. Harvest. This module would depend on what crops were growing and if they can be mechanically harvested like grains or potatoes, or have to be manually picked like tomatoes.

Off the shelf parts of course would be preferred As ChristopherN mentioned a ATV frame/power train are easy to come by and fit into the right size, The transmission would have to be geared lower than stock and a PTO added but those problems have been already solved. Control systems and sensor arrays are going to be the hard part. A Agricultural bot is not going to work a 30,000 acre farm by itself..But a 160 acre farm with a few humans around No problem.

Thanks Mike

Edited 1 time(s). Last edit at 08/24/2014 05:00PM by Miertam.

Your idea is great indeed! However, if the project will focus on third world country. We should consider how many farmers use smartphone? The condition of electricity as well as their internet provider. Since most of rural areas in third world country have this kind of problem.

I too live in the Midwest and thought about such a system.
What I envisioned was a skid loader sized vehicle, with multiple attachments.
My design had thoughts of battery powered hydraulic pumps. Swap-able battery packs that were wind powered/solar recharged(charging station). The bot would autonomously change battery packs as needed.
GPS controlled areas, with some sort of safety shutoff.
Some sort of remote alert system when something breaks (cell phone enabled was my pick).
the list goes on...

I paid no attention to cost while I thought about things.

The idea was to go small, but it can run 24 hours a day.

The idea came from another project I was working on which was a small vegetable garden tiller to keep the weeds down.

Edited 1 time(s). Last edit at 09/05/2014 10:59PM by cat.farmer.

If you are unfamiliar with Open Source Ecology, you should definintely check them out. Their Tractor is well along in development, and I'm sure they could use another designer!

As a midwesterner located in Oklahoma, I'm interested.

I've been wanting to do a farm bot for a long time now and would like to collaborate.