Robots and farming

be afraid be very afraid a land without people coming soon http://www.theguardian.com/australia-news/2016/jun/04/transforming-the-bush-robots-drones-and-cows-that-milk-themselves
Just say no to robots

David

Ultimately it will be too expensive and will fail, in my opinion.

It's inevitable. There's so much to be gained by using new technologies to make the most of the land we have.

A network of sensor spikes providing real-time temperature, light, and moisture data, all connected wirelessly, could provide an in-depth map of microclimates for use in planting any site. By planting only crop guilds suitable to specific microclimactic conditions, crop health can be optimised, and losses decreased.

Shock-deterrent collars, or hopefully something sound and smell-based as an ethical deterrent tuned to natural animal cues, with changeable boundaries controlled wirelessly and using GPS references could enable mob-grazing on whatever scale, and allow ranchers to move their herds to fresh grazing multiple times a day, mapped over time and able to be monitored over a smartphone. This would free the rancher/farmer/homesteader for other tasks.

And how intelligent would it be to ignore tools that could help us be better stewards? Better information enables better decision-making. Assisted monitoring frees up time for other tasks; the computers do the observing, and we make the conclusions.

Chris Kott wrote:It's inevitable. There's so much to be gained by using new technologies to make the most of the land we have.
A network of sensor spikes providing real-time temperature, light, and moisture data, all connected wirelessly, could provide an in-depth map of microclimates for use in planting any site. By planting only crop guilds suitable to specific microclimactic conditions, crop health can be optimised, and losses decreased.

Shock-deterrent collars, or hopefully something sound and smell-based as an ethical deterrent tuned to natural animal cues, with changeable boundaries controlled wirelessly and using GPS references could enable mob-grazing on whatever scale, and allow ranchers to move their herds to fresh grazing multiple times a day, mapped over time and able to be monitored over a smartphone.

I'm not seeing any robots in that scenario.

I imagine a fruit harvesting robot. Some sort of mechanical sloth, that can slowly climb around every crevice, harvesting small berries that would otherwise be uneconomic to harvest. Most berries change color when they are ripe, and that is something that could easily be detected.

Here's a list of quite prolific berries that don't lend themselves well to current forms of mechanical harvest. Because of this, vast quantities go unpicked , while people go to the grocery store and buy inferior food . I'm sure the list is much larger. These are just a few that are abundant near me. Autumn olives, salal berries, huckleberries, Himalayan blackberries, wild strawberries. And the list goes on.

I would gladly program my solar-powered sloth, to gather up every wild strawberry at my place. They are much better than the store-bought variety, but also quite small. If I were harvesting them for sale, I would have to charge $50 for one of those small half pint baskets. I'm willing to pay a couple thousand dollars for this sloth, should anyone here feel up to the task.

A fruit-picking robot would have to be very sophisticated, with the sensitivity of human eyes and fingers!

Here it comes:    

Most likely to become available to only large corporate farms, in my opinion, possibly contributing to putting more small farms out of business.  Mid-sized farms might be able to afford them if they become as common other industrial robots.

"Complete with controllers and teach pendants, new industrial robotics cost from $50,000 to $80,000. Once application-specific peripherals are added, the robot system costs anywhere from $100,000 to $150,000. Reconditioned robots are a less expensive option. Typically, used robots cost half as much as new robots."  https://www.robots.com/faq/show/how-much-do-industrial-robots-cost

Cost of common industrial robots seem competitive with the cost of other large farm machinery.  Initially of course, fruit picking robots will be very expensive.

We have a lot of "robots" on the dairy farms here and have for several years now. These are not big factory farms either, but family dairy farms, maybe 300-500 cows, a few family members working the farm, etc.

Most of these "robots" end up being for the calves and heifers since they cannot drink the milk off their mothers (since that is going in the milk tank). So these "robots" mix and warm the milk, then dispense it, but only the right amount, and at the right time. It does this by reading the tag reader on the calf or heifer. When a calf approaches the feeder and is not suppose to get a drink...it gets a squirt of water it its face deterring it. No shocj collar needed.

The end result has been phenomenal, about a 25% reduction in mortality just because the animals are getting exactly what they need.

>>>

Tractors are essentially robots as well. With GPS they really do not need a farmer in the cab. Even the GPS and controls go beyond the tractor, but to the implement itself like planters and crop heads. I know on our John Deere corn planter, it used to be when you came to a place you already planted at an angle, there was two choices:(1) pick the whole planter up at once before you reached where you planted, or (2) over-plant through where you already planted and then pick the planter up. Today EACH corn row picks up so there is no overseeding, causing too high density stocking in the field and wasting expensive corn seed. It has to be seen to be appreciated.

>>>

Robots that milk cows, robots that grain cows based on their lactation and individual needs, choppers that send how many tons is coming out of the spout to the computer at the office automatically...the list goes on. Robots are not GOING to arrive on family farms, they are already here, and have been here for awhile now.

Chris Kott wrote:
It's inevitable. There's so much to be gained by using new technologies to make the most of the land we have.
A network of sensor spikes providing real-time temperature, light, and moisture data, all connected wirelessly, could provide an in-depth map of microclimates for use in planting any site. By planting only crop guilds suitable to specific microclimactic conditions, crop health can be optimised, and losses decreased.

Shock-deterrent collars, or hopefully something sound and smell-based as an ethical deterrent tuned to natural animal cues, with changeable boundaries controlled wirelessly and using GPS references could enable mob-grazing on whatever scale, and allow ranchers to move their herds to fresh grazing multiple times a day, mapped over time and able to be monitored over a smartphone.


I'm not seeing any robots in that scenario.

I was generalising. They don't need to be human-form, or even a single discrete unit. The word "robot" comes to us from the Czech writer Karel Capek in his 1920 story R.U.R. I think it's derived from the word for work, as the root is similar to the Polish for work (roboty, if I remember the spelling). The key here is the doing of work, not the fitting in with science fiction tropes or industrial expectations. A camera drone can, if set to run a program, say, taking detailed footage in real-time of pasture to monitor grazing and regrowth, easily be seen as a robot by this definition. I would also accept the abovementioned dongle and sprayer system for the feeding of calves, or for the automated milking of cows, as one that is inherently robotic in nature.

And if we can get away from the HAL comparisons, I could see a computer mind collating data from sensor nodes placed at regular intervals all over any property to provide a map showing everything a gardener, farmer, rancher, or homesteader could possibly want, updating in real-time and storing data over time. Think of the detailed observations that could take place! It could be programmed with basic horticultural knowledge, enough to be able to make recommendations, based on what will thrive where, of whole lists of plants that you could guild up for specific microclimates. It could even be designed to account for known plant functions, such that you will be alerted if a guild lacks a function, sufficient nitrogen-fixers or nutrient accumulators, for instance. Such a system could learn about your preferences based on historical decision making, and eventually make suggestions that take into account individual tastes.

And the reason this will likely happen is the same one that tractors have gone the way of automation: cost savings. We can actually benefit more from some of the automation technology being made today, because we aren't using sensors, say, to inform us about where we need to water, or spray something, or whatever. As permaculturalists, we can actually design sophisticated tiny ecosystems in each of our different microclimates. We can have dozens of nanobiomes, should we choose to look at it that way. Observation and information give us that power, as much as anything else.

-CK