5 ms·

Ask HN: Would you automate 100 acres of farmland

I have the opportunity to draft a proposal to farm 100 acres of urban farmland. It is iffy financially mostly because of labour availability. And costs. But. I am still going to go ahead and create a proposal for a working farm ..a kind that has never existed. An automated urban farm with a kitchen-sink/duct-tape approach wrt available Agtech. Even if it gets rejected, it will become part of public records(it is public land). I am ok with it. Maybe someone will take it up some day and run with it even if I can’t now...but perhaps it might go through..regardless.. I am familiar with how the previous farmer farmed it as he has been mentoring me for the past 6 something years with my own farm. I want to take all that and modernize it. 1. Starting with minimal and automated tillage. 2. Swarming robots for cultivation, weed management and data collection of pest/inputs/yield prediction. 3. Updated paperwork and certification. I want to create my own protocols, compliance and audit standards for automated farms. Including traceability and best practices. Bad news: it’s super cost prohibitive. Because I have to be grounded in reality, I am going to chisel it to suit the project. Having said that...I can’t help fantasize about cutting edge futuristic farm of the future..with existing Agtech. Any ideas? Forget about the costs. Just vision.

rini17 6y ago
There is http://agrokruh.sk/ http://agrokruh.sk/ (sorry could not quickly find english version). Instead of robotic swarms that would be difficult to maintain, it uses simple rotating arm with farm implements attached as needed. It is independent from fossil fuels, runs silently on modest amount of electricity. The emphasis is not on cutting-edge tech but on self-sufficiency and on improving the quality of the soil.
- jelliclesfarm 6y ago
  That’s a good idea. There are many ways automation can help. Data, self sufficiency, labour, water, environment etc. for me..I want to focus on small urban Ag and labour saving tech automation. Our challenge here in the states ..California..is labour. Labour availability and affordability. I also envision that plant breeding and hybrids will be integral to any kind of farm automation. I am already working with a couple of plant breeders on this. But only for a few main crops. Lots of work to do.
  - rini17 6y ago
    Full automation in the open field is bound to fail as there are so many factors you can't predict, weather, pests, market demand. You will inevitably end up with some kind of warehouse hydroponics under artificial light. In urban environment there is always (at least leisure) labour available. You have to stop fixating on labour as a problem and instead think how to attract people to come and maintain the garden.
    - jelliclesfarm 6y ago
      We do not have labour. Those who do not actively farm for their living do not understand how serious the labour issue is.. 100 acres is not a ‘garden’. The denial of labour availability as a problem by non farmers and tech sector is why there is no labour automating innovation in Ag. Why isn’t anyone listening to us?
      - giantg2 6y ago
        People are listening. Look at Ag Funder. They invest in agricultural tech. There have been advances in automating aspects of agriculture. Tractors that can drive themselves, irrigation systems based on soil moisture, etc. One of the problems is that small farms are less and less viable since the 80s and have been consolidated under large companies. There's not much incentive to develop solutions to small farm issues when the big farms source labor for a couple cents per pound. Even when the solutions are available, they will price out the small farms.
        jelliclesfarm 6y ago
        Ok. Thanks for your input.
      - rini17 6y ago
        I am not listening to you because you seem to insist on automating 100 acres of land in urban environment. This implies you have to keep out all the people living around, to prevent them from interfering with the automation. That sounds utterly absurd. Why not doing it on large scale (to reap automation scaling advantage) and away from cities (to leave more urban environment for... how to say it... human activity?) Oh, and 100 acres _is_ a community garden. In Europe these are everywhere.
ed_db 6y ago
Satellite and Drone imagery.
giantg2 6y ago
Your best bet is to plant blackberries and semi-dwarf fruit trees. Then open it as a pick-your-own place. You would need minimal labor.
- jelliclesfarm 6y ago
  Harvest would be brutal. Can’t automate that!
  - giantg2 6y ago
    That's why you crowdsource as a pick-your-own.
    - jelliclesfarm 6y ago
      That is not a viable business model. High foot traffic and liability issues. Would a person running a factory allow random people to enter and do part of the work for DIY inside their premises? Just because Ag happens under the open skies doesn’t mean that it’s a park with free entry to all. Working farms are professional spaces and are for profit.
      - giantg2 6y ago
        LOL. Pick-your-own is a common model found at farms. Do some Google searches.
        jelliclesfarm 6y ago
        With all due respect..I don’t know what your domain expertise is, but I am not going to ask you to ‘do some Google searches’ for you do your job. This is about automation and latest tech enhanced mechanization of small farms. Not how to sell produce.
        giantg2 6y ago
        Actually the premise of this post is about writing a proposal to farm 100 acres of urban land. Automation could help in that endeavor and was listed as one of the aspects that could be explored. Often the best solution is the simplest. Pick-your-own is a common model used by small farms to reduce labor requirements for some types of produce such as apples, strawberries, etc. This approach would also require less capital investment than specialized automated machines. There's nothing wrong with googling aspects of your job - how else will you discover new things if you don't seek them out. In fact, most software devs research job related information via Google searches all the time.
pretzell 6y ago
I just got hired on a ~100 acre farm to basically do this. We have one other person who's been working on a auto weed puller (autonomous driving over rows and computer vision to pick weeds). Would love to talk more about this! my email is zach_asmith@yahoo.com
- pretzell 6y ago
  It also looks like you're in the bay? We're in San Gregorio CA
- jelliclesfarm 6y ago
  Thanks! Will reach out! Yes..the farm is actually 136 acres right in the middle of Bay Area!
eb0la 6y ago
Forget robots. Gather data on how the plans are doing (think lora-wan sensors) and try to minimize both waste and costly investments (like unneeded fertilizers and treatments). Hire a someone who knows how plants grow and when they are in their better condition. Surprisingly some plants give the sweetest fruits when they are near death (think grapes and wine). When you have data, you can focus on investment and ROI. It is also more prudent ;-)
- eb0la 6y ago
  Do not forget to have a small part of the plants as a control group. I mean, plant that don't get the special treatment (fertilizer/medicine/etx?). If you don't do that, you're buying snake oil in less that three years
- jelliclesfarm 6y ago
  We have a labour problem. There aren’t many people who are willing to do repetitive manual labour for minimum wage. We need robots. Plants grow without data. Farming needs manual labour and mechanization(that requires skilled labour). We need to automate farming.

rini17 6y ago

There is http://agrokruh.sk/ http://agrokruh.sk/ (sorry could not quickly find english version). Instead of robotic swarms that would be difficult to maintain, it uses simple rotating arm with farm implements attached as needed. It is independent from fossil fuels, runs silently on modest amount of electricity. The emphasis is not on cutting-edge tech but on self-sufficiency and on improving the quality of the soil.

jelliclesfarm 6y ago
That’s a good idea. There are many ways automation can help. Data, self sufficiency, labour, water, environment etc. for me..I want to focus on small urban Ag and labour saving tech automation. Our challenge here in the states ..California..is labour. Labour availability and affordability. I also envision that plant breeding and hybrids will be integral to any kind of farm automation. I am already working with a couple of plant breeders on this. But only for a few main crops. Lots of work to do.
- rini17 6y ago
  Full automation in the open field is bound to fail as there are so many factors you can't predict, weather, pests, market demand. You will inevitably end up with some kind of warehouse hydroponics under artificial light. In urban environment there is always (at least leisure) labour available. You have to stop fixating on labour as a problem and instead think how to attract people to come and maintain the garden.
  - jelliclesfarm 6y ago
    We do not have labour. Those who do not actively farm for their living do not understand how serious the labour issue is.. 100 acres is not a ‘garden’. The denial of labour availability as a problem by non farmers and tech sector is why there is no labour automating innovation in Ag. Why isn’t anyone listening to us?
    - giantg2 6y ago
      People are listening. Look at Ag Funder. They invest in agricultural tech. There have been advances in automating aspects of agriculture. Tractors that can drive themselves, irrigation systems based on soil moisture, etc. One of the problems is that small farms are less and less viable since the 80s and have been consolidated under large companies. There's not much incentive to develop solutions to small farm issues when the big farms source labor for a couple cents per pound. Even when the solutions are available, they will price out the small farms.
      - jelliclesfarm 6y ago
        Ok. Thanks for your input.
    - rini17 6y ago
      I am not listening to you because you seem to insist on automating 100 acres of land in urban environment. This implies you have to keep out all the people living around, to prevent them from interfering with the automation. That sounds utterly absurd. Why not doing it on large scale (to reap automation scaling advantage) and away from cities (to leave more urban environment for... how to say it... human activity?) Oh, and 100 acres _is_ a community garden. In Europe these are everywhere.

ed_db 6y ago

Satellite and Drone imagery.

giantg2 6y ago

Your best bet is to plant blackberries and semi-dwarf fruit trees. Then open it as a pick-your-own place. You would need minimal labor.

jelliclesfarm 6y ago
Harvest would be brutal. Can’t automate that!
- giantg2 6y ago
  That's why you crowdsource as a pick-your-own.
  - jelliclesfarm 6y ago
    That is not a viable business model. High foot traffic and liability issues. Would a person running a factory allow random people to enter and do part of the work for DIY inside their premises? Just because Ag happens under the open skies doesn’t mean that it’s a park with free entry to all. Working farms are professional spaces and are for profit.
    - giantg2 6y ago
      LOL. Pick-your-own is a common model found at farms. Do some Google searches.
      - jelliclesfarm 6y ago
        With all due respect..I don’t know what your domain expertise is, but I am not going to ask you to ‘do some Google searches’ for you do your job. This is about automation and latest tech enhanced mechanization of small farms. Not how to sell produce.
        giantg2 6y ago
        Actually the premise of this post is about writing a proposal to farm 100 acres of urban land. Automation could help in that endeavor and was listed as one of the aspects that could be explored. Often the best solution is the simplest. Pick-your-own is a common model used by small farms to reduce labor requirements for some types of produce such as apples, strawberries, etc. This approach would also require less capital investment than specialized automated machines. There's nothing wrong with googling aspects of your job - how else will you discover new things if you don't seek them out. In fact, most software devs research job related information via Google searches all the time.

pretzell 6y ago

I just got hired on a ~100 acre farm to basically do this. We have one other person who's been working on a auto weed puller (autonomous driving over rows and computer vision to pick weeds). Would love to talk more about this! my email is zach_asmith@yahoo.com

pretzell 6y ago
It also looks like you're in the bay? We're in San Gregorio CA
jelliclesfarm 6y ago
Thanks! Will reach out! Yes..the farm is actually 136 acres right in the middle of Bay Area!

eb0la 6y ago

Forget robots. Gather data on how the plans are doing (think lora-wan sensors) and try to minimize both waste and costly investments (like unneeded fertilizers and treatments). Hire a someone who knows how plants grow and when they are in their better condition. Surprisingly some plants give the sweetest fruits when they are near death (think grapes and wine). When you have data, you can focus on investment and ROI. It is also more prudent ;-)

eb0la 6y ago
Do not forget to have a small part of the plants as a control group. I mean, plant that don't get the special treatment (fertilizer/medicine/etx?). If you don't do that, you're buying snake oil in less that three years
jelliclesfarm 6y ago
We have a labour problem. There aren’t many people who are willing to do repetitive manual labour for minimum wage. We need robots. Plants grow without data. Farming needs manual labour and mechanization(that requires skilled labour). We need to automate farming.