You might wonder why I want to turn land management over to robots. Is it because I’m such a geek that I think everything goes better with robots? No, not really. Sure, I think the technology is cool, but I’m not eager to factor human beings altogether out of any activity, not even those that are dull, dirty, and/or dangerous.

I am, however, eager to see the benefits of replacing methods designed to spread a human operator’s time as thinly as possible with methods which reintroduce attention to detail to plant cultivation. Granted, that attention would, for the most part, be provided by robotic sensors, processors, and algorithms, but that has an upside as well as a downside.

The downside is mainly that it’s unfamiliar and we don’t know whether we can trust it. The upside is that robots aren’t limited to human senses, but can go as far beyond them as available technology will support, providing information not directly available to us, and they also aren’t limited to the time a human operator can afford to invest, provided they are capable of autonomous operation. Potentially this means that instead of a single machine running 40-60 hours per week, you can have many machines running 24/7, vastly increasing the available machine time per land area.

Combine that rich sensory information and machine time availability with manipulators capable of performing basic gardening operations (planting, weeding, pruning, harvesting, saving seed, …) and you have a platform justifying the effort to develop higher level software which supervises beneficial plant combinations in space (polyculture) and time (crop rotation), which is capable of mixing annuals with perennials (permaculture), which makes room for native plants (particularly those that are threatened or endangered), which also makes room for animals without allowing them to ruin crops, and which can make sure there are enough flowers, throughout the season, to keep pollinators healthy.

Such a system could also learn from experience, even setting up intentional experiments, maintain the genetic diversity of crops, and accelerate the creation of new strains in response to changing conditions. It could also provide exquisitely detailed information in support of operation management.

Technology of this sort would make for a far more varied and interesting landscape (biodiversity) and greater variety of production (diversified income), supporting a more well balanced diet for those dependent upon it. It could also dramatically reduce the energy and other inputs used in crop production, all but eliminate soil erosion and the airborne dust and stream contamination that result, reverse the loss of organic matter in the soil (improving its water absorption and retention and contributing to the sequestration of atmospheric carbon), save many endangered species from the brink of extinction, and bring the practice of agriculture into parity with other aspects of the modern world in its sophistication and the respect it commands, helping to enhance rural culture.

So, yes, that would be very cool technology, but it’s really about all of the benefits that could be realized through its widespread deployment.

We industrialized our agricultural firms, consolidating them into multinational, billion dollar publicly traded corporations. As a result, they no longer have nutritional responsibility, they have fiduciary responsibility. And that fiduciary responsibility makes them not create healthy calories, but cheap, popular ones.

LED lights are far more efficient than incandescents, but enough light to drive photosynthesis will still require significant power, so one still might ask where the power is going to come from.

In case you’ve forgotten, this video provides a clear explanation of the concept behind and the mechanical workings of the TE+ND Rover.

For more, check out their page on the competition website. You can vote using either a Facebook ID or email address.

Oh, by the way, Prospero is also a contestant.

The first video in this playlist is a presentation given last year at the announcement of the event. The rest were taken at the event itself, and show the nature of the competition as well as something of the level of sophistication of the competitors.

University of Colorado Assistant Professor Nikolaus Correll, speaking at TEDx Front Range earlier this year, outlined his vision for transforming agriculture through robotic cultivation.

(Correll Lab at the University of Colorado)

This will be the 10th edition of the Field Robot Event. Organized by Fontys University of Applied Sciences and Wageningen UR (University & Research), it will be held in Venlo, The Netherlands, on the grounds of Floriade 2012.

(PDF of slides from above presentation video about the 2012 Field Robot Event)

More information available here and here.

Here’s just a few examples: ALPHA Evo, AMAZONE Pantera, AGCO SpraCoupe, and a selection of sprayers from Agrifac.

Note that these machines are not built for traction, but to suspend a load high enough that it doesn’t interfere with crops. This and the adjustable track width is why, taken together as a class of machines, they make a good model for a robotics platform.

Ideally, only the frame and running gear from one of these designs would be used. To that add a transverse beam (gantry) along which suspended robotic modules can move sideways, mounted far enough above the ground surface to make room for the modules and to avoid interference with the crops.

The application of field robotics to agriculture is an emerging area of interest for our researchers. The increasing demands on our agricultural sector are forcing farmers to consider robotic assistance where before they worked alone. In recent years GPS guided tractors have become commercially available and are now seen commonly in many countries in the world. These systems still rely on the farmer to supervise them – normally from tractor’s cab. It is hoped that the next generation of farm robots will be more aware of their immediate surroundings and will be capable of mapping obstacles and navigating autonomously. Unlike field robotics in other domains such as mining or the military (where safety and the removal of people from hazardous situations is a major driver), agricultural robotics will only make sense when the business case means that using robots will save money when compared to farming in a traditional manner.

The concept of Integrated Pest Management has been around for several decades, and is just as useful in a robotic, no-till scenario as it is combined with conventional, traction-based agriculture, perhaps even more so. In a nutshell, it’s about the inclusion of biological controls in a mix of methods that can also include mechanical and chemical controls. The use of barn owls in this video is an excellent example.

Corn (maize) is a good example of a crop for which these conventional measures of productivity tell a story of brilliant progress, with the result that corn is cheap enough to use not only as livestock feed, to be converted into meat and dairy products, but as the feedstock for production of ethanol for fuel, competing with fuels refined from petroleum pumped from the ground, rather remarkable considering that corn kernels represent only a small fraction of the biomass of a corn plant and that fermentation and distillation aren’t particularly efficient processes.

Crops that fair less well by these measures include many vegetables and most fruits, which have been becoming gradually more expensive, especially as compared with grains that are easily handled mechanically, but even compared with meat and dairy products from grain-fed livestock. One major consequence of this has been that people generally consume more grains, meat, and dairy products, and less fruit and vegetables than they once did, before the mechanization juggernaut got started and while vegetable gardens were still common.

So, by an altogether different measure, how healthy the average diet is, mechanization has been a disaster, so far. I say “so far” because the essential problem is that, so far, mechanization has favored crops consisting of hard, dry seeds, that are easily handled in bulk, making other crops needed for a balanced diet relatively less affordable. In happier economic times this would matter less, as people would simply pay the premium for a healthier diet, but the times being what they are people are scrimping however they can, including with the food they consume.

There are other ways of measuring productivity: energy use*, soil gain or loss*, water use and contamination*, and the degree to which a given practice denies space to native flora and habitat to native fauna. By any of these measures, conventional mechanization comes out looking at least shortsighted if not dimwitted. *(per unit produced)

So is the answer to turn back the clock on agricultural technology, to replace the plow with the hoe and the drill with the planting stick? I’m not prepared to make that argument – although I’ve no doubt others would – aside from noting that gardens are a better use of many urban spaces than are lawns, and there is no further need for rural communities to supply cities with cheap labor, since those cities are already well supplied, and many rural areas suffer from depopulation.

Instead, my position is that we need to take mechanization to the next level, replacing dumb machines suited only to bulk operations with smart machines capable of performing well-informed, detailed manipulations, for example controlling weeds by selectively pulling them from the ground or pest caterpillars by picking them from plants (unless they’ve already been parasitized, as by wasps) rather than by applying poisons.

Given machinery with an adequate array of sensors and a sufficiently broad range of optional actions, applying best practices becomes a matter of mating these with processing power connected to an expert system, and of programming.

It gets better, because the same system that works the land can be used to improve the expert system through experimentation and, in routine operation, by accumulation of data to which statistical methods can be applied, and can also be used to improve the crops themselves, as for instance by leaving the best formed, most insect resistant cabbages to go to seed.

The bottom line is that this approach can make available the mechanical equivalent of an attentive expert gardener, at a cost, given predictable economies of scale, that would make possible the wholesale replacement of conventional, traction-based machinery and methods with more adaptable machinery bringing a whole new repertoire of methods to bear, one far better suited to the production of the fruits and vegetables that have been becoming unaffordable under the current regime.

As for the other measures of productivity mentioned above, such machinery, since it wouldn’t need to turn soil in bulk and could operate long hours without continuous supervision, would consume energy at a relatively low rate, suitable for supply from solar panels or via the grid from renewable sources. It could operate through continuous ground cover, all but eliminating soil loss, and with minimal use or complete non-use of herbicides and pesticides, reducing soil and water contamination. Ground cover, mulch, and the humus accumulating from decaying roots can also reduce the need for irrigation, and the ability to create local varieties through seed selection based on the health of maturing plants can further reduce it, as well as helping to adapt more quickly to climate change. Making room for native species, something that can only be accomplished in conventional practice by leaving land completely undisturbed, becomes a matter of programming the system to leave certain species alone, wherever it finds them, even to the extent of tolerating some crop loss to native fauna, and to leave anything it can’t identify alone until it can be identified.

Such machinery might not be able to compete with conventional practice in the production of corn and other bulk commodities, at least to start with, but it also wouldn’t consume prodigious amounts of petroleum-based fuels. Moreover, development and rapid deployment of such machinery would drive the growth of a new, potentially domestic industry, one that would also work to the benefit of materials recycling efforts, more efficient transportation, and on and on.

The R-word I haven’t yet mentioned is robotics. While such machines probably aren’t what most people first think of when robots are mentioned, their creation and production falls squarely within the discipline of robotics, composed as they would necessarily be from robotic technologies.

Reposted from Lacy Ice + Heat, via Cultibotics.

Besides the usual gardening techniques, such machines could move some soil and gravel from the rail-bed around to create microclimates with various shading/exposure, slope, drainage, water collection and/or even distribution across a flat bottom. They could weave vines together to create sheltered spaces for small birds, train high branches of trees from both sides of the rail-bed to arch overhead, creating deep shade by tying them together, position art objects intended to provide habitat for mice, birds, and bats and anchor them with soil and gravel. These machines could also assist other species in the creation of their preferred shelters, for instance by digging a bit of a hollow at the bases of trees with roots that spread abruptly just under the surface, or providing platforms in the forks of tree branches, just big enough to support proper nests, constructed of sticks and twine.

Emphasis on avian habitat would mean faster accumulation of a diversity of plant species, because birds frequently pass the seeds of berries they’ve eaten through, undigested. And, because the seeds of berries preferred by birds predominate, the result is a positive feedback loop.

Such machines could also provide damage-free access to the resulting space through inclusion of observation decks on top of the robotic rail platforms. If several such machines are to be spread along a single rail-bed, they should be designed so that they are able to approach each other closely enough that their observation decks come together, allowing riders to step across from one to another.

While I fully expect Emma Marris would deem the use of robots to create her Rambunctious Garden a distraction, and my contention that robotics is necessary to the achievement of sustainability non-obvious, at best, I still must consider her an ally, for preaching the gospel of making the best of what we have left.

My hope is that the goal of rewilding can be interwoven with that of producing food, fiber, and fuel, on a grand scale, through the creation and application of robotic systems of land management capable of nurturing crop plants through their entire life cycles while leaving neighboring plants undisturbed.

I firmly believe that (short of convincing the vast majority of people to return to subsistence farming, something which could only be accomplished through intense coercion) robotics is vitally important to achieving sustainability.

Realizing others’ mileage might vary, I took that post to a conferencing system I’ve been on for over twenty years and asked whether the notion seemed counterintuitive to anyone there. Unsurprisingly, perhaps, it did. In fact, I was probably the only participant in the conversation for whom the idea wasn’t at least a bit odd.

Understand that we’re not talking about your standard social network fare, here. The other participants in the conversation are, to a person, all intelligent and (otherwise) well informed.

Seeing that the conversation had pretty well run its course, I concluded with the following:

It’s unfortunate that so much of robotics is weapons research, and even more unfortunate that the associations most people have with robots and robotics is of clunky machines that are unintentionally dangerous. The clunkiness is a passing phase, and already an anachronism in many cases, but I can see why some prefer to avoid the word “robot”, talking instead about intelligent or adaptive machines. In Japan they speak of RTs, Robotic Technologies, which makes for a nice refocusing in my humble opinion. Robotic technologies find there way into all sorts of objects not usually considered robots. A more general restatement of the proposition I laid out [here] would be that robotic technologies are essential to the achievement of sustainability. This might be an easier sell, however I really do mean robots, complete with interchangeable manipulators on the ends of arms with at least a few degrees of freedom, and operation that is sufficiently autonomous to break the 1-to-1 correspondence between machine and operator, with the machines conceivably running 24/7 during the busiest season (and maybe drawing some power from the grid to keep them running through the night). It’s my belief that the use of such machines is the only way we’ll ever manage to bring best practices to the vast majority of land in production, and that the best that is possible without them will prove unsustainable in the long run.

That this point of view was at least initially counterintuitive for the unusually astute social environment in which I posed it means to me that there’s still a great deal of work to be done to repair the perceptual damage done by the preponderance of robot portrayals in fiction and to jumpstart creative imagination for how autonomous machinery might help us surmount the difficult challenges before us.

If your goal is to maximize the production of those crops that are easily produced and handled in bulk and survive long-term storage well, in the interest of generating return on capital investment and foreign exchange, and only care about how it’s done insofar as that impacts the bottom line, you might conclude that capital expenditures to further minimize payroll would generally not be cost effective, that it would cost more to replace the remaining workforce than to keep it.

However, if you’re interested in guaranteeing the sustainability of production far into the future, despite climate change, while also halting soil loss, ending the use of poisons, preserving remaining diversity in both crop and native genomes, and rebalancing production for healthier diets, you may need both more sophisticated machinery and all the people you can recruit.

Such a complicated goal implies complex operations, and complex operations imply a large variety of tasks, some easily mechanized and others common enough to make mechanization worthwhile, even though challenging. Those that are neither common nor easily mechanized will fall to human workers, farmers and farmhands, who are far more adaptable than any machine.

At some point in the future it may become possible to build machines adaptable enough to take the place of a farmer, but until the annual cost of ownership of such a machine drops below the annual cost of one human worker, it won’t make economic sense to deploy them, and without an infrastructure to drive down the cost of robotics, that may never happen.

I had a pretty good idea of what computing was about from an introduction to CS class in which we wrote FORTRAN programs on cardpunches. At that scale there was no help to be found from that direction, but the advent of the microprocessor changed everything. Suddenly it became thinkable to have mobile devices each with its own electronic brain. My mind reeled with the possibilities, but there were a million unknowns.

One thing was clear, though, if Moore’s Law was even close to being correct it wouldn’t be long before the speed of the electronics was no longer the hangup. It would be the mechanical designs, the software, much of which would depend on transforming biological knowledge into computer code, and the chicken/egg problem of creating an industry and a market for that industry’s products at the same time.

And that’s pretty much where we are now. The speed of the electronics has so far exceeded the other pieces of the puzzle that even if we might wish for still more it’s a moot point. We’re not putting what’s available to good use.

Remember, we’re talking here about getting what we need from the land while honoring the back-to-the-land aesthetic of living lightly upon it, as a species, but not about people fleeing the cities to scratch out their personal livelihoods with whatever meager assemblage of skills they might manage to collect. That could be more destructive than factory farms.

The solution, really the only possible solution if we’re to stop soil erosion, ground water and stream contamination, the loss of biodiversity, and the gutting of rural culture, is robots. That’s right, robots.

Only by substituting machines which can be invested with some understanding of ecology – or which are at least well suited to play a role in an ecologically sound approach – for the dumb machines currently in use, can we have it all, our comfortable lives, a reliable supply of food of varied types, and a clear conscience.

I’d love to be telling you about all of the cool developments in cultivation robotics, how this team had succeeded in building a system that could differentiate between closely related species immediately upon sprouting, and how another had created a tiny robot that ran on the body fluids of the aphids it consumed. I wish I could report that the USDA had funded research into intermingling rare and endangered native species with crop species and making room for moderate wildlife populations without sacrificing too much commercial productivity. Heh, at least I can truthfully say it could happen, which seemed pretty far fetched just one year ago.

Realistically, though, nearly all of that sort of work remains to be done, and it’ll be a great ride when it finally does begin to happen!

Harvest Automation didn’t start out with the idea of building robots for greenhouse operations, instead they looked around for a market where their initial efforts, to develop what was essentially a larger, more powerful version of the Roomba (without the vacuum), would be applicable outside of the initial context, instead of their having to start from a blank slate for each application.

They settled on agriculture, beginning with a machine to move potted plants around, operating alongside human workers. It isn’t hard to imagine how this platform might develop in various directions to perform other horticultural tasks, eventually evolving into a scalable system capable of applying intensive methods to large land areas. That’s not the stated goal of Harvest Automation, but the profit motive may very well lead them in that direction.

This video consists of the earliest participants in the Arduino project talking about how it got started.

Checking Vimeo, there are over 4,000 videos there tagged “arduino”. [2012Oct14: now over 9,000]

Similarly, a search of YouTube with “arduino” as the search term results in over 5,000 hits. [2012Oct14: now over 100,000]

Found on Engadget.

Reposted from Cultibotics.

The participating agencies are:

• National Institutes of Health (NIH)
• U.S. Department of Defense (DOD), (http://www.defenselink.mil/)
• National Science Foundation (NSF) (http://www.nsf.gov/)
• United States Department of Agriculture (USDA)
• Department of Homeland Security (DHS)

Each agency’s interest in the initiative is described within the document.

I noted the presence of USDA on the list with extreme interest, as you might expect, and am pleased to report that an effort to develop the sort of system I’ve previously described (replacing traction with dextrous manipulation) should be fundable within their guidelines.

Found on Danger Room.

Reposted from Cultibotics.

My interest is in improving agricultural practice, and I think robotics presents the approach most likely to serve that end, really the only approach with any chance of widespread success. (For me, robotic tractors are merely annoying, except as they help generate experience with autonomous navigation in an uncontrolled environment, applicable to other systems.)

Conversely, agriculture may be the largest potential market for robotics [in terms of revenue], one so large that it could drive the development of self-reconfiguring, self-reproducing robotic factories. This depends on the total cost of operation using robotic devices coming in below the total cost of operation using conventional methods, which includes increasingly expensive fuel for tractors (which might be replaced by solar-generated electricity in the robotic scenario).

I’m very encouraged to see robotics finally gathering momentum, and have hope that some of that momentum will find its way towards radically transforming agriculture.

Reposted from Cultibotics.

This can be done using pots of various sizes on a platform, repotting plants as necessary. It might also be done using a grid or honeycomb-like support frame, each cell of which is large enough to accommodate a single mature plant of the largest variety to be grown this way, but which is also divisible into smaller cells – rectangular in the case of a grid, or a combination of hexagonal and triangular subsections in the case of a honeycomb – for seedlings and smaller plants.

This approach, because it would mean discrete positioning, would lend itself to automation. It would also position the soil surface at the same level for all plants, rather than having smaller pots hidden and shaded by larger pots. While something resembling repotting would still be needed, because a suspension system can have a soft underside, such as a loosely woven fabric pouch, made of biodegradable fiber, hung from a rigid frame, that repotting could be nothing more traumatic for the plant than positioning a smaller frame within a larger one and filling in between with potting soil, leaving the pouch in place to decompose while the plant’s roots grow through it, a procedure which could be accomplished robotically, without the need for high precision. This can be repeated until the stem of the plant grows to the point that it no longer fits through the smallest subframe initially employed, which usually [wouldn’t] happen.

Like pots on a platform, when a plant is removed from the framework, the soil is typically removed with it, which can help with the control of pests and diseases. (Used soil, containing whatever is left of the pouches, which can simply be cut loose from the frames pieces, can be sanitized by inclusion in compost, which can hold a temperature between 120 and 160 degrees F, for several days, the peak temperature depending on the scale of the compost operation as well as on the initial ingredients and how it’s managed.)

Reposted from Cultibotics.

Willow Garage is out to make it happen in robotics.

They not only have developed a robot, the PR2, with sufficient dexterity to fold towels, and given eleven of these to research institutions around the world, but they’ve also released “an open-source, meta-operating system” for robots, called ROS, which you can download and use for free.

This is a great beginning, and could prove to be the seed of what develops into a standard platform.

Reposted from Cultibotics.

Reposted from Cultibotics.

This ability to get custom parts economically means that individuals and small companies can go where only large companies could realistically go before, and should infuse new energy into the culture of small scale experimentation, which was already showing renewed signs of life over the last decade or so.

This is very hopeful!

Reposted from Cultibotics.

Applying the notion of elegance to machine behavior may resonate for some and not for others. What could it possibly mean, elegant machine behavior, wouldn’t that be a contradiction in terms?

In this piece on another blog, I suggest that Apple should get into robotics, partly because to fail to do so would be to leave the largest looming growth market to others, and partly because I believe the company has something to contribute, something relating to elegance. I think Apple would set a high standard for machine behavior, and then exceed it, providing a tangible example of first-order elegance.

I say “first-order elegance” to suggest that there is also a “second-order” or “meta-elegance” that looks beyond present behavior to its ultimate effects. For example, formality may appear elegant, but if children are subjected to it all the time they may fail to develop emotional intelligence, an inelegant result.

As applied here, it is second-order or meta-elegance that is more important. It matters far less whether machines that tend land appear deft in their actions than whether the result of those actions appears more garden or desert-like. That’s not to say that first-order elegance is unimportant. Efficient movement, of the entire machine and of its parts, is an important aspect of cost-effectiveness, but efficiently producing a undesirable result gains nothing.

I believe that second-order elegance is achievable in this context, that machines can be programmed to understand complex living systems and nurture them, while raising food and fiber for market in their midst. If I didn’t believe that I would never have bothered trying to explain this vision of a greener future founded on robotics.

Reposted from Cultibotics.

The second is the combination of a robotic hand possessing touch sensitivity and quick, flexible movement with a fast vision system, allowing some rather amazing manipulations of objects (check out the video!).

Of the two, the latter provides me far more hope for the future of robotic land management. A pair of hands like that, mounted on comparably quick arms, themselves mounted on a mobile platform, could be expected to cover every square foot of a several acre plot, every day, performing mechanical operations like planting, weeding, pruning, and harvesting. This represents a significant head start on the necessary hardware.

It’s becoming clear that the hardware development will pretty much take care of itself, as basic abilities like this are developed and combined. The software may require more focused effort; probably will.

Reposted from Cultibotics.

Reposted from Cultibotics.

Wheat comes in two main growing patterns, winter and spring. Winter wheat is planted in early fall, where winters are relatively mild, grows to a lawn-like few inches in height before winter, and then resumes growing as winter recedes, and is ready for harvest in late spring or early summer. Spring wheat is planted in late winter or early spring and is ready for harvest in late summer or early fall. Either way the ground looks much like a disaster zone between harvest and replanting, especially with old-school tillage which begins with plowing under the stubble from the last crop, and, like most disaster zones, it’s a prolific source of dust. To live in wheat country is to live with a landscape in this mutilated condition several months out of each year.

The fault isn’t so much with the technology in use (aside from the choice of tillage regimes), but is rather a result of monoculture, the planting of a single crop year after year. Even using standard tillage practices, something as simple as a crop rotation system might have wheat harvest immediately followed by the planting of something else, clover for instance. However, much of wheat country gets insufficient rainfall to support more than one crop per year. In fact, without irrigation, one crop every two years is more common on the high plains, meaning that even during the height of growing season half the land under cultivation continues to look desolate and contribute to the dust-load in the lower atmosphere, at the cost of some of its own fertility.

Nothing about the equipment in common use precludes crop rotations, and row cropping systems can manage two, or even three crops in the ground at the same time, but three is about the limit. Using conventional methods, intermixing a dozen or more species, other than for pasture or hay, is unthinkable, no matter what benefit might result. So is continuous cropping unthinkable. You can grow pumpkins among the corn stalks in the fall and snow peas climbing up them in the early spring, but sooner or later you’ll need to turn under the debris, if not to prepare a seed bed for more corn then to keep the thistles in check. [Using conventional methods you can’t simply uproot the corn stalks and gather up the pumpkin and pea vines and toss it all on the compost pile, at least not working around corn sprouts that were planted through the debris.]

For continuous cropping, you need more deft handling of soil and plant materials than implements pulled by tractors can provide. You need something more like what a gardener does.

With continuous cropping there’s always something in the ground to break the wind and keep down the dust, and while the field will never have the look of ripe crop of monoculture wheat, all ready for harvest at once, it will also never look like a desert, nor, with proper handling, like a thistle patch.

What you get is a landscape that’s more varied throughout the year, but not so starkly punctuated by season.

What you also get is more variety in production. Instead of wheat, wheat, and more wheat you might also get squash, beans, onions, peppers, millet, and so forth, as well as perennials like sand plums, apricots, currents, and mulberries, all the makings of a healthy diet.

The key to making this possible is dexterity combined with attention to detail, such as could only, until recently, be supplied by people. The key to making it practical is robotics.

To the extent there is any close connection between the quality of land management and the quality of life that land supports, it follows that the quality of life achievable through the best available land management method will be better than what can be achieved without it.

You cannot economically duplicate, by any other means, the quality of land management that is achievable through the appropriate application of robotics.

Reposted from Cultibotics.

Ahem! Farming has been conducted primarily by machines for going on a hundred years, at least in the United States. I want to substitute intelligent machines that proceed carefully and work continuously for big, dumb machines that are designed to get the job over with as quickly as possible, and at the same time move farmers out of their roles as machine operators and into the roles of technician and manager.

The threshold that still looms large but is shrinking with every passing week, primarily due to advancements in military robotics, is autonomous operation, making it possible for one person to manage many machines simultaneously, instead of being symbiotically fused to one for the entire time it is in operation.

At some point in the not too distant future, it will become practical to turn tractors loose under robotic control, but by the time that happens that same threshold will already have been crossed by less powerful, less dangerous machines. Moreover, once tractors arrive at autonomous operation there’s not much to drive further development. Sure, you can push efficiency higher and accident rates lower, but it’s still the same old thing.

With the sort of detail-oriented systems I’ve been attempting to imagine and describe, that threshold of autonomous operation is just the beginning, the spark that lights the rocket. Knowledge that would be of no use to autonomous tractors – because they’d still just be pulling implements around a field – could improve the performance of machines using a horticultural approach and improve the productivity of land they tend. Moreover, they would be able to discern much of that knowledge for themselves, through experience (statistics applied to crop measurements) and sharing information with each other.

For every increment in sensory capability, processing power, mechanical versatility, and software sophistication there would be a potential payoff, in machine performance, productivity, and/or the quality of the overall result.

That’s what a growth market looks like.

Reposted from Cultibotics.

An alternative to the above would be a four fingered manipulator with basal segments angled at less than 90 degrees, possibly as little as 45 degrees, with space between them, and the ability to rotate into that space. In the case of 45 degree angled basal segments, each finger should be able to rotate 22.5 degrees in either direction from its rest position. This would result in both a manipulator with two sets of opposed digits, 45 degrees apart on either side, arranged along either of two axes, as well as a more effective sheer. [2012Oct13: This allows switching the plane of the gripper formed by the fingers 90 degrees through finger rotation alone, no wrist rotation needed.]

The basal segments to which I’ve been referring would appear, from anywhere along the central axis, like simple blocks of metal presenting a distinct edge. However, from the outside, with those edges hidden, they would look more like the links in multi-jointed digits that they would also be.

[2012Oct13: To be quite honest, I’ve never been able to imagine exactly how the basal sheer would work in practice, how the parts would connect and be activated. This aspect now seems far less useful than the ability to regroup fingers to form grippers, switching between perpendicular planes, by rotating each finger 45 degrees. For pruning, you’d want a gripper next to a sheer, with the whole assembly being rotatable, so you could hold onto the branch you were removing while cutting it loose, and then pull it free.]

Reposted from Cultibotics.

One likely solution is to surround the lens with a doughnut-shaped (toroidal) trough, with an outer opening at least as large as the lens itself. Compressed, filtered air would be introduced tangentially, through a jet positioned along the trough’s widest circumference, causing the air within to spin. The rate of spin would increase as the air climbed the inner wall of the trough toward the opening, and continue to increase approaching the axis of the lens.

There would also be a net movement of air outward through the opening. Once outside the confines of the trough, air would spray outward in an approximate disk shape. Any dust that happened to come close to the opening would be pushed off by this disk-shaped, laminar outflow, with essentially no chance that it would end up inside where it might come in contact with the lens.

Reposted from Cultibotics.

But considering only the labor component for the moment, if a machine is capable of working at approximately the same speed as a human worker, but can do so continuously, day and night, assuming the human is working a forty hour week, the machine can perform the work of four human workers, with eight hours per week left over for maintenance. If one person, working forty hours per week, can intensively manage one hectare, then a machine working at the same speed can manage four hectares (nearly ten acres) with the same degree of intensity.

If and when cultibots begin to become common, the comparison with a human performing the same sort of work will become less relevant, and the capacity of such a machine will most likely be measured in terms of the amount of land it can effectively manage according to a particular pattern, which requires a large subset of the range of operations such a machine might be able to perform. (As with any benchmark, such measurements will favor machines that are optimized for them.) And the value of that work will be the value of the produce at market, minus the cost of transportation to get it there, which, as now, will be largely a matter of what the human owner chooses to grow, and whether the weather works for or against them.

Given that such a machine might also be programmed to make a little room for native plants and animals, and that doing so would constitute a social benefit, the value of the work done might include modest subsidies for the land set aside in this way, probably in return for verifiable data that could be spot-checked for accuracy. At some point, once agricultural practices based on traction (pulling implements across the surface of a field) had become uncommon, the subsidies might be replaced with a simple requirement that a percentage of productive land, distributed in a manner designed to favor threatened and endangered species, be made available for native habitat. At that point, those still using tractors would need to bring in surveyors to stack off the parts of their fields they couldn’t till, whereas those using cultibots would simply need to include the proper programming.

Reposted from Cultibotics.

On the other hand, greenhouses are most useful when combined with regular gardens and used seasonally to start plants earlier than they could be started in the open, or when the shade from other plants would impede sprouting or development. Even a relatively frail machine, better adapted to spending its time indoors, might venture out in calm weather, long enough to set out plants it had started in trays and peat pots, provided it was sufficiently mobile.

This scenario, a machine that does the tedious work of planting seeds and tending plants in a greenhouse, moving them out to open ground when conditions allow, is what might be termed a natural starting point for the development of such machines, a more limited, more surmountable engineering problem than a machine intended to perform all aspects of land management. A machine applied this way need not be able to perform absolutely every horticultural operation to be useful, nor would it need to be able to deal with a completely uncontrolled environment.

It’s very likely that there are other such natural starting points for the development of cultibots. Collectively, these natural starting points represent a threshold of minimal investment before a return on that investment can be forthcoming. Once that threshold has been crossed, at any point, incremental improvements should be adequate to insure that machines which can handle the whole job are eventually produced, and the return on that initial investment could be very sweet indeed.

There’s another, equally important threshold to consider, the automation of the production of these machines. So long as they are hand-crafted prototypes, they have no chance of competing economically with hand labor, or, as is more likely, with the transportation of produce from milder climates. Mass production will get them into the game.

Self-reconfiguring factories that not only build such machines but which can also replicate, by building the equipment for new factories and the machines to assemble them, will drive down the cost to the point where the logic behind it all becomes inexorable, but by that point we’re no longer talking only about machines for land management, and there had best be very solid safeguards in place. The point of mentioning this scenario in the context of cultibotics at all is that land management may be the one application of robotics where the size of the potential market could justify the investment to cross this threshold. Once crossed, of course, the technology would be generally applicable.

Reposted from Cultibotics.

To begin with, while you can see the potential for it in what exists today, the technology largely remains to be developed, so figure five years of R&D and experimental installations before anything starts rolling off an assembly line, and probably another five to get the bugs out to the point where it’s really possible to let the machines run without some degree of supervision.

At that point, ten years hence, you might still have to drive twenty miles [two hundred?] to see one of the new machines in operation. Then, for at least another ten the story would be one of them becoming very gradually more common, as well as more sophisticated. Meanwhile there’s mouths to feed, hundreds of millions of them, and business as usual will necessarily continue.

At some point, maybe twenty-five or thirty years out, the size of the market for food cultivated by autonomous machines would surpass the size of the market for conventionally grown food, and at about that time I would expect to see several things happen. For one thing, the largest tractors would disappear from the market, as there would no longer be sufficient demand to justify their production. Also, a shakeout would begin among tractor and implement companies that hadn’t gotten into robotics themselves. On the other hand, the infrastructure for getting grain and produce to market could be expected to improve, under pressure from [and borrowing techniques from] robotic operations with their more diverse output and their ability to provide detailed information about what they would need to move how soon.

Some crops, however, would continue to be more economically produced by conventional methods. In particular, it would be difficult for generalized machines using horticultural methods to match the efficiency of traction-based monoculture in the production of small, dry-land grains – wheat, barley, oats, and rye. At least in the near term, the reduction in acreage dedicated to raising these crops by conventional methods would result not from direct robotic competition, but from the substitution of more fruits and vegetables in place of grains in the diets of both people and livestock.

Granted, with the advances in robotics that all of this activity would bring about, the tractors on the market then would likely also be capable of autonomous operation, although I’d expect to see a lot of hold-out farmers, using older equipment, still spending long hours driving their tractors, powered by synthetic fuels. Whether that practice will ever become as uncommon as farming with horses had become in 1960 is anyone’s guess.

Reposted from Cultibotics.

Similarly, if you want to build a multi-vendor market for almost anything, one of the best things you can do is to find the natural divisions of responsibility and insert standard interfaces into the boundaries between them.

One example, in the context of cultibotics, would be the connections between robotic arms and tool units that attach to them. What physical form should the connections take? How much force should the mechanical connection be able to withstand or apply? What services should the unit be able to expect from the arm? What signals should each understand or send to the other, or pass through to the CPU? Would the arm supply water, or should any unit making use of it have a hose connected to it in addition to its connection to the arm?

Detailed answers to these questions would fill a thick book, which is what it typically takes to specify a standard. Moreover, chances that any standards organization which undertook to fill in the details would decide that there would need to be several such standards, to accommodate scales ranging from very small to very large.

But given a set of well-defined standards, you’d be able to buy a robotic arm from company A and a tool unit complying to the same standard from company B, and have good reason for confidence that you could just plug them together and have it work seamlessly.

Until now there hasn’t been much need for standardization in agriculture. The prime examples of what there has been would be power takeoff, hitches, and hydraulic connectors, all of which have been standardized by the ISO, which makes it the most likely candidate for tackling standards for robotics in agriculture.

Of course the ISO isn’t going to get involved until there’s at least the beginnings of a market and more activity than sparse experimentation, so it behooves those who do get involved early to cooperate with each other to develop ad hoc standards which are in the public domain, royalty-free, or available for low-cost-per-unit licensing, suitable to the bootstrap nature of the field. These ad hoc standards can later serve as the starting point for formal standards.

Reposted from Cultibotics.

What would the ideas encapsulated here look like if embraced by the Obama-Biden team? What might they be called? Here’s a few focused statements that occur to me…

• Help insulate farmers and farming regions from dependency on volatile bulk commodity markets by encouraging greater diversity of production.
• Facilitate production improvements through both simultaneous and sequential polyculture.
• Enable farmers to grow more of their own food without need for much time investment or manual labor.
• Reduce the time spent in machine operation.
• Reduce the acreage needed for an economically viable farming operation.
• Reduce the initial investment required to start a farm.
• Provide farmers and their children with high-tech experience.
• Create a demand for skilled technicians and technical instructors in rural areas.
• Create opportunities for rural youth.
• Preserve local crop varieties and experiment with new crops.
• Improve the quality and diversity of locally available produce.
• Reverse the impoverishment of rural culture.
• Reduce exposure to pesticides and pesticide residues.
• Reduce the dependency of agriculture on fossil fuels and feed stocks.
• Reduce contamination of runoff and ground water.
• Reduce and eventually reverse the loss of soil fertility.
• Reduce wind-borne dust.
• Enlist productive land in the efforts to preserve endangered species and provide wildlife habitat.

This list could be far longer, but that should be enough for a sample.

Reposted from Cultibotics.

First, these machines will necessarily have sensory components. Digital cameras and microphones are practically a given, but they may also have infrared imaging, radar and/or laser scanning, chemical sensors to provide something akin to a sense of smell, pressure/stress sensors for a sense of touch, probes for soil moisture, temperature, pH, O2 content, and nutrient availability, weather instruments, and some means of locating themselves very precisely relative to the boundaries of a field or other stationary reference. Compared to most machines, they will have available a rich collection of information about their environments, rich compared even with what human senses provide.

Next, they will have significant computer processing power, sufficient to take the data streams from all of these sensory devices, find patterns in them, compare them with each other and with historical data (including the exact position of every seed and when it was planted), create and update a real time 3-dimensional model of their immediate surroundings, locate items of interest within that model, choose a course of action, and send the detailed instructions to the machine’s moving parts, closely monitoring their progress.

Finally, they will have various moving parts, likely including high resolution or specialized sensory components that can be sent in for a closer look. Those moving parts might include a range of grips, from fine tweezers to something strong enough to uproot small trees, mechanical snips, lasers with enough power to fry a meristem, high-pressure water jets capable of slicing through the stem of a plant, fingers to move other plant material out of the way, a vacuum for sampling air at ground level or removing insects, sprinklers and sprayers, trowels of various sizes, and, of course, the soil probes mentioned earlier. Such tools might be combined into sets incorporated into units which could be plugged onto the ends of articulated arms and quickly switched out.

That’s a basic outline, but we need to return to the data processing hardware and the code it runs to fill out the picture, since it can make the difference between an expensive toy and a productive machine that more than pays for itself. A major task the processor must perform is resource scheduling, and to do that effectively it must sort actions into those that can be performed without moving anything massive (slow) and without switching out tool units, those which require either movement or a tool switch but must nevertheless be accomplished before moving on, those which can be left until a future pass over the same area but not indefinitely, and those which can be left undone unless it becomes convenient to do them. Efficient scheduling also means mapping the movement of even the smallest parts so they proceed smoothly from one thing to the next, without having to retrace their paths more than is unavoidable.

An important point to be taken away from the previous paragraph is that scrimping on computing hardware and software is likely to prove counterproductive, by reducing the overall capacity of the machine disproportionately. We should expect the computing components to represent a substantial fraction of the overall cost of the machine, and we shouldn’t be surprised if they also consume a substantial fraction of its energy budget. Better to invest an extra 10-20% to make a given physical machine capable of performing the work of two, and to invest 1 or 2 kilowatt-hours to save ten.

Something which should be apparent from this mental exercise as a whole is that what’s being proposed is largely a simple extrapolation of technologies which already exist. There are already mechanical arms and mechanical grips; there are already sensors and various means of controlling machine operation. What’s mainly missing is the software which would turn data streams into a 3D model in a horticultural context, choose what to do, schedule resources, and map out the details. That’s a lot left to be done, requiring a significant investment for a long term payoff, but it’s a fairly straightforward problem, and divisible into more manageable chunks. Let’s get to it!

Reposted from Cultibotics.

There’s no single, right answer to these questions. Rather there’s a wide range of potential answers, some of which will likely prove more workable than others. Let’s look at some of the possibilities.

What would it look like? Almost anything, from a snake-like device slithering along the surface, to what appears to be little more than a single wheel rolling about, to a platform supported by long, spider-like legs, to a beam supported at each end by wheeled trucks. It may turn out that the best arrangement is a mixture of larger and smaller machines, with the larger ones designed to never put their weight on soil being cultivated.

How would it be powered, and how would it transmit power to the parts that need it? They could get their power directly from the grid, from engine-driven generators, from wind generators, from photovoltaic panels, from concentrating solar collectors, or simply from batteries or other energy storage. Any such machine will need at least a small amount of electricity, to power the electronics. Mechanical power could also be electrical, but needn’t be. It might be provided via compressed air. Delicate, articulated parts might be moved via fine cables or wires, much as our own fingers are moved by tendons linking to muscles in our forearms.

What actions would it be capable of performing? Planting seeds, of course, beyond that the possibilities are nearly endless, but even the placement of seeds can be accomplished in many ways.

In conventional agriculture, seeds are typically inserted into the soil in rows, through an opening created by a disk (a rotary knife), and covered over by a roller. This is an efficient method of planting a large area to the same crop, or even to a mixture of crops with seeds of approximately the same size, if you don’t mind running the planting device and the tractor pulling it over the same soil surface through which the seeds will have to sprout and in which the sprouts will have to grow. Most such planting devices require the soil first be prepared into a seedbed, meaning that plant debris from previous crops must either be turned under, with a plow, or broken down by a combination of tillage, decay, and weathering, to form a relatively uniform surface, easily broken into small particles. A few such planting devices are capable of placing seeds through rough plant debris, rendering the preparation of a seedbed unnecessary, but they’re still mainly used to sow a single crop to a large area.

A robotic gardener would also need to be able to place seeds not only through plant debris but between standing plants. It would do so one at a time, perhaps very rapidly, like a sewing machine, but still one at a time, and far more precisely than any bulk planter, positioning them in the most advantageous microenvironments available. Even when planting in bulk, the use of rows would be optional, and in many cases a honeycomb-like pattern might prove preferable.

Weeding might be accomplished by identifying weed seedlings and removing them while still in the sprout stage. Undesirable plants that propagate by root spreading could be controlled by injecting steam below the surface, wherever they appeared. Insects, like aphids, could be controlled by removing infested leaves. Diseases could be controlled by removing infected plants. Nutrient deficiencies could be identified early and treated quickly. If plant debris needed to be reduced to mulch it could be clipped off at ground level and shredded, without disturbing the soil.

In each case the action taken would be local and specific, rather than applied to an entire field, and generally would not involve moving large amounts of soil around, a practice which wastes both energy and soil fertility.

But the essential requirement, without which this whole scenario would be futile and meaningless, is that the machines must operate autonomously, puttering through their days without constant human supervision. They must have both the ability and the latitude to choose what to do next for themselves. Considering they must also operate in uncontrolled environments, this is the greatest challenge.

Reposted from Cultibotics.

While requiring more energy than tillage-free management would, this system has the advantage of operating directly on electricity, meaning that there’s no necessary dependency on petroleum to keep it running. It has two electric motors, which appear to be in the 2-3 horsepower range. Using both, its maximum power requirement should be no higher than 5 KW. Using just the motor that drives the wheel at the end of the rotating beam, which would be the more common case, the maximum power requirement should be no higher than 2.5 KW.

It also has the advantage of being usable now. In fact, to judge by the video, it appears to already have been in regular use for several seasons.

Reposted from Cultibotics.

The best example of which I’m aware, Robert Silverberg’s The World Inside, describes a world divided between urban towers and the land between them. The land between is tended by machines which are themselves tended by people, a rural population with a very different culture from that found in the urban towers.

While the world Silverberg describes is more of a dystopia than a utopia, not least because it is fast approaching limits that it steadfastly denies, that aspect of the book, the use of intelligent machines to enable a superior grade of land management than could be achieved without them, rings true.

Reading that book was most likely the beginning of my own obsession with the subject, although I don’t clearly remember how it started.

Reposted from Cultibotics.

Clearly I badly misestimated some factor: the rate of progress in computing and robotics, the difficulty of adapting these technologies to the array of tasks involved, the tremendous momentum of business as usual once it becomes a matter of money rather than simply conventional practice, or the degree to which others might share the vision that was burning in my brain.

If anything, I underestimated the rate of progress in computing, which has been going gangbusters since the invention of the personal computer, almost without a break. On the other hand, I probably overestimated the rate of progress in robotics. There’s been quite a bit, but we’re not yet to the point where you can assemble a complete machine for practically any purpose from readily available parts in stock. That time is still coming, but it isn’t here yet.

I think I’m a realist about business momentum. I understand the deep conservatism that guides most investment, and the heavily conventional nature of most marketing types, allowing for only incremental change. No surprises there.

What’s left is the difficulty of adapting technology and the degree to which others might comprehend and share my vision, two factors I believe to be connected in a sort of chicken/egg (which comes first?) relationship.

It’s easier to imagine a field being managed by robotic machinery if you have an example of such a machine sitting in front of you. On the other hand it’s easier to think about building such a machine if you have a clear idea of what all it’s supposed to be able to do, and how fast it will have to work to succeed, with what safeguards – basic design parameters.

Lacking the means to contribute much in the way of machine design, I’ve concentrated on elaborating and pushing the vision, hoping others better positioned to work on the hardware might become interested.

[2012Oct13: The final portion of this post removed for being what I’d term navel-gazing.]

Reposted from Cultibotics.

This arrangement allows the positioning of the active components over any portion of a strip of land as wide as the machine’s span and of indefinite length, while confining its wheels to narrow, widely spaced tracks, which can be graveled to mitigate damage.

Aside from the suspended components, it might be thought of as a vehicle with very wide track and relatively short wheelbase.

While such a machine might weigh as much as a conventional tractor, that weight would largely be relatively inexpensive extruded or rolled materials, welded together, as opposed to cast or forged parts with machined surfaces.

Reposted from Cultibotics.

In the context of agriculture and the potential for applying robotic technology to its improvement, this principle applies at least two ways.

The first of these, because it is happening regardless of anything else, is how agriculture has been changing over recent decades, and how the human culture of farming, rural society and the rural landscape, the robustness or fragility of the crops themselves, soil fertility, and biodiversity have changed as a result.

This is something of a mixed bag. For example, on the one hand you have a proliferation of poisonous substances used to control various pest species, but on the other you have the growing popularity of Integrated Pest Management, which uses them sparingly. And while some of the practices which became ubiquitous in the wake of the dust bowl have since become less common, the rising cost of fuel works in favor of lighter tillage, leaving some stubble, which helps control erosion.

Nevertheless, this situation looks rather bleak overall, particularly given the heavy dependence of agriculture on fuel and other products derived from petroleum, and a comparison between ourselves and the frog in the gradually warming water is a bit too apropos.

The other way in which the slow accumulation of change applies is in the development of the various tools and technologies needed for robotics in general to flourish. The array of what’s available for use is already good and getting better, if not quite rapidly then at least inexorably, and the more complete the toolkit the more applications become economically feasible, further accelerating the pace of development. At some point that logic is bound to take hold, powerfully and irreversibly, if not this year then maybe next year, or the year after, or maybe it’s already begun and just happening slowly enough that it’s hard to see.

Here too the boiling frog applies, in that the robotics industry could flourish without contributing anything significant to the improvement of agriculture. It could simply fail to live up to that particular potential, there being no shortage of other, more clearly profitable potentials to be chased after, and plenty of encouragement from DARPA with regard to military applications.

Here we are in the pot, with the general state of agriculture growing ever more tenuous and the industry with the power to transform that situation taking no notice, much less recognizing its all-important role.

[2012Oct13: I now know that the situation wasn’t quite as bleak as it seemed to me at the time, and that there are quite a few individual roboticists who are well aware of the potential robotics holds at least for facilitating horticulture, and a handful of robotics programs doing practical work in this direction.]

Reposted from Cultibotics.

Fast forward to 2008.

Take your standard recommendations as to what constitutes a balanced diet; work up a meal plan for a week, and from that a shopping list; go to any supermarket and price out your shopping list. You’ll find that some items, basically those that can be grown and harvested without the use of hand cultivation, are relatively inexpensive, and others, those requiring manual labor for at least one step in the process of getting the crop to market, are relatively more expensive. It’s all too tempting to just go for the less expensive items and leave out the more expensive items, maybe using vitamin supplements to make up for what’s missing, maybe not.

This is a hidden cost of current agricultural practice, that it makes a nutrition-poor survival diet relatively inexpensive, while a really balanced diet is unaffordable to many.

Intensive cultivation using robotic land management could do a lot to make currently expensive produce, and therefore a balanced diet, more affordable.

Reposted from Cultibotics.

Granted that it’s far easier to create such a program for a machine intended to vend beverage cans than for one intended to plant, tend, and harvest a mixture of crops in an uncontrolled, nonuniform environment. The latter is harder, but not impossible, nor even so far beyond some other tasks to which computers have already been put, such as handling a complex mix of financial transactions or handling the control surfaces of an otherwise unstable aircraft.

Plants have numerous characteristics which, if not already known, are at least measurable. Soils likewise can be measured and typed. Weather is the most unpredictable factor, but even weather is regular enough to be categorized as a climate, and statistical approaches allow decisions to be made in the face of unknowables, which on average work out well enough. It’s a complex context, but not unmanageably so.

Thankfully, it isn’t necessary that any such system perform perfectly. It’s only necessary, as an initial condition, that it either perform on par with conventional practice while costing less, or outperform conventional practice while costing about the same.

This becomes easier with each passing year, as the cost of diesel fuel and other petroleum-based agricultural inputs steadily rises, and as the power of computing equipment and the sophistication of programming tools improve.

The rising cost of petroleum works against conventional practice and for a dramatically different approach utilizing robotics because conventional practice is utterly dependent upon huge amounts of fuel, fertilizer, and pesticides, whereas robotics can make possible an alternative approach without those dependencies.

The alternative approach itself is as old as gardening. Robotics can make this approach a contender by duplicating the knowledge, skills, and hands of the gardener many times over, making it applicable to land areas measured in hundreds or thousands of acres instead of square feet. That’s where the distinction between a robot and other types of machines comes in. A robot is a machine with a brain of its own. Robots can operate autonomously, meaning there doesn’t have to be a human operator present and you don’t need as many human operators as you have machines.

That’s an essential point. It’s the one-to-one correspondence between operator and machine that has shaped conventional practice, exerting upward pressure on the size and power of the machines as economics exerted upward pressure on the amount of land required to earn a decent income. And large, powerful machines are only really useful for conducting operations that are applied uniformly over entire fields. The result is mile after mile of fields, all planted to the same crop at very nearly the same time.

Breaking that one-to-one correspondence is the key to breaking the dependence on oil, while at the same time dramatically increasing the variety of production and relieving the oppressive monotony that has so overtaken the rural American landscape.

And the key to breaking that correspondence is machines with enough brains to operate autonomously.

Reposted from Cultibotics.

Without a transformational vision, more than parsimonious transformation is unlikely to result.

I aim to provide such a vision, and to help connect and empower those who find themselves enlivened by it.

This vision is about fusing ecological reasoning into every aspect of the technology employed to conduct land management, specifically including the management of land engaged in agricultural production.

That creating such a technology is possible is beyond reasonable doubt, but whether the resources to do so will be brought to bear in a timely manner, and whether that technology will find a clear path to widespread adoption, are far less certain.

It may require that we demand [or conjure] it into existence.

Reposted from Cultibotics.

Machines can make use of senses we don’t possess or which are more sensitive than those we do. Their vision can extend into the infrared and ultraviolet, as well as more finely dividing the visible spectrum, and can also be more detailed and quicker (tracking faster motion) or more accurately track changes over a period of days or weeks. Their hearing can be far sharper than our own. They can be equipped with chemical sensitivity capable of distinguishing between substances we would group together under broad categories, like sweet or acrid. They can also be equipped with radar and sonar, laser ranging and scanning, accurate measures of temperature, humidity, and insolation, and their manipulators can be made to gauge and control pressure more accurately than do our own fingertips. In short, machines can have far better data available to them than would an unassisted human gardener in the same position.

Machines can also correlate information very quickly, drawing on recorded data and expert systems to make decisions, and applying heuristics to experience to refine those expert systems. A machine might reasonably be expected to identify to species every plant within the area it was tending, to know whether they were considered crops, benign, weeds, or threatened or endangered, and treat them accordingly. It might be expected to predict to an accuracy of a few days when it could harvest a particular crop, and estimate to within a few percentage points the quantity that could be expected, barring a calamity such as hail or a tornado. It might also be expected to adapt a cropping plan to market conditions, for example putting in more of some crop that hadn’t done well elsewhere and would therefore be in demand.

Machines can whisper to each other, via radio links, over distances far greater than a human shout will carry. They can coordinate their activities precisely, cooperating toward a common goal without so much as a hiccup.

Machines can, as has recently been demonstrated by DARPA’s autonomous vehicle competitions, operate in an uncontrolled environment.

The foregoing is intended as a glimpse of how it might work once development was far along. It presumes a mature technology, some of the pieces of which aren’t yet available or only just beginning to be so.

Reposted from Cultibotics.

Maybe it moves on little wheels, maybe like a snake, maybe like a centipede. Maybe it has low-set, horizontal snips in place of a mouth, as for shearing off crabgrass every time it grows back. Maybe it uses its tail like an ovipositor, to plant seeds. Maybe it produces a loud hiss to scare off hungry rabbits and deer.

Probably it would keep tabs on soil moisture and possess an array of chemical sensitivities for detecting various soil conditions, like whether anaerobic decomposition is happening near the surface.

Such a device might also apply dusts or mists to the underside of leaves, for example to control fungal growth with a light application of copper salts, or caterpillars with spores of a particular strain of bacteria.

Whatever the details, there’s a place for smaller devices, representing considerably smaller investments, in a mix of machines that collaborate to manage productive land.

Reposted from Cultibotics.

Before responding to that question, let me turn it around. Why, given that there are no laws preventing them from doing so, are so few people in the ‘modern world’ engaged in by-hand cultivation, or even cultivation using draft animals?

One big reason is that land isn’t free, and the cheapest land is located far from city markets, as well as from the amenities cities provide. And, even if there were still land for homesteading, there’s the matter of potable water, housing, tools, seed, food enough to last until the first crops come in, canning supplies and equipment if part of that crop has to last through the winter, and some means of transporting a portion of the crop to market. If you’re growing one of the local commodity crops, you can almost leave off the transportation part, since there will already be a network in place for that purpose, and you’ll only need to move your crop as far as the nearest node in that network, but you’ll also be competing with other producers benefitting from economies of scale for the thin margins they’re accustomed to. If, on the other hand, you’re growing watermelons in wheat country, you’ll have to provide your own transportation.

Okay, back to the original question, why robots, given that there are people in need of work?

People are slow. One person can just about manage five acres by hand, if they work at it full time. Depending on the climate and what’s being grown, that might be enough to feed as many as twenty, or at least provide them with vegetables, which isn’t such a bad consumer/producer ratio, but it’s far lower than the average for conventional agriculture. One person using the sort of equipment you see everywhere in rural America, can work between 600 and 1,000 acres, planted to crops like wheat or soybeans that lend themselves to bulk handling.

People are expensive. Who’s going to pay the unemployed to produce by hand a small fraction of what is already being produced far more cheaply by means of machines under direct human control? Even if the current cost of production were to rise several fold, current practice would still result in cheaper food than what workers earning minimum wage could produce by hand. Granted that the economics are better for crops that don’t lend themselves to linear mechanization (performing field operations while/by moving over the surface), and that a diet limited to those that do would be significantly impoverished.

People are lazy have other priorities. And well they should! (Personally, I’m a fan of Bucky Fuller’s idea of paying people, modestly, to go to school. They benefit by becoming more marketable, and society benefits through an increase in the skill-level of its workforce.)

People would [still] be needed. A conversion from tractors to small, autonomous machines [applying horticultural techniques] almost certainly would not result in a net displacement of workers, rather the reverse, and you’re more likely to find enough people (4..8, at a guess) willing to maintain (and augment*) the machines that manage a square mile of land than you are to find enough people (about 100) willing to work that land by hand.

*(Especially at first, it’s unlikely that robots would be able to do everything that needed to be done, and the gaps in their capabilities would have to be supplied otherwise, most likely by people.)

Reposted from Cultibotics.

For example, if these robots that I’ve been talking about aren’t engaged in tillage, what are they doing? That remains an open question, since there are undoubtedly useful techniques I haven’t yet thought of, but, for an idea of what might be possible, consider what gardeners can accomplish with their own two hands and short-handled tools. That’s the scale of manipulation I have in mind, working with individual plants and the spaces into which they’re to be inserted.

Would such robots have human-like hands? [Probably] only in the vaguest sense; they’re likely to have manipulators with opposable, finger-like appendages. Would they stir the soil like a gardener does with a trowel? Maybe. Would they use something like snips to do pruning? Probably, although there might be a better approach to pruning than mechanical snips, like a high velocity water jet (such as are used to cut steel in some industrial settings).

It isn’t necessary, nor even desirable, to exactly replicate the set of techniques used by a gardener. Such machines would need a repertoire of techniques sufficient to manage a garden, but while some of their techniques might seem quite familiar, others might be quite beyond the capability of a human gardener.

For example, if a machine were able to identify a weed seedling early enough, it need only destroy the seedling’s meristem to interrupt the growth of a weed. This requires very little energy, and might be accomplished by a precisely targeted, high velocity water droplet [or flechette-shaped bit of ice, or even a pulsed laser]. Using this method, a machine might deal with several weed seedlings per second, limited only by the speed with which it could identify them and reorient the nozzle [or mirror], all without any disruption to surrounding plants.

More tenacious weeds that sprout from roots could be pulled out, except that they sometimes break off just below the soil surface, and their roots may pass below plants you’d rather not disturb. An option would be steam injection, through a tube inserted next to the stem. Another option would be coring, removing a cylinder of soil around the stem to a depth of a few inches. Yet another option would be to use electrical current to heat the weed. These are all techniques that a gardener might use, but, except for grabbing ahold of the base of the stem and pulling the plant out, they aren’t common.

Compared with weeding, seed planting would be relatively simple. On the other hand, transplanting seedlings started elsewhere would be more challenging, although mechanical systems for this purpose probably already exist and could be used as a model [and the use of compressed peat or compost pots, which are left in place to enrich the soil, would simplify the process].

Dealing with mid-season issues, like insects and nematodes, microbial infections, plant nutrient deficiencies, and so forth, is hugely complicated, and will require considerable development effort. But small-scale machines have an advantage in that they can deal very specifically with the effected leaf, plant, or location, and also in that, because they would revisit each location frequently, they should be able to catch problems early.

Harvest is also somewhat complicated, since each crop type presents its own set of challenges. What works for wheat doesn’t work so well for maize. What works for tomatoes won’t be sufficient for pumpkins. [Specialized] hardware attachments may be needed in some cases.

This vision isn’t a fantasy, but there’s a lot of work to be done.

Reposted from Cultibotics.

In this effort to spread human involvement over more area, the first thing out the window is any operation which can’t be performed linearly, like plowing, by moving through the field along rows, and the main casualty of this limitation is intensive intercropping. You can mix two or maybe even three crops together in alternate rows in the same field, but not twenty, like you might in your garden; it’s just too cumbersome. It’s really more convenient to plant just one crop at a time, and as we move upscale, from oxen to tractors pulling implements a hundred feet wide, that convenience becomes a matter of practical necessity, and even a minor lack of uniformity in the land itself becomes an annoyance. The result is flattened fields planted to a single crop, as far as the eye can see, and travelers on any highway passing through it hurrying along because it’s so boring.

Attention isn’t the only thing being extracted in this scenario, so is soil fertility. Tillage, something we’ve taken for granted for ten thousand years, means unnatural aeration of the soil, which in turn means rapid oxidation of its organic content. Single-cropping means long months of exposure to wind with minimal cover or none at all. The net effect is called “desertification”.

So what does all this have to do with robots, and how might they be part of a solution rather than simply making the problem worse?

The simplest definition of a robot is a machine that responds to its environment. The nature of that response might seem trivial in many cases, as in determining the exact position of the tip of a welding rod in relation to the parts being welded and adjusting accordingly, but it’s a start, and, as robotic technology advances, more sophisticated responses become possible.

It’s unfortunate that the same word, when applied to a human, means exactly the opposite. For this reason I generally refer to “robotics” rather than “robots” to make it more clear that I’m talking about technology in which the acquisition of information and its use in determining the behavior of the machine are essential characteristics, in many cases the most important ones.

It’s also unfortunate that “robot” overlaps so much with “android”, which refers to a human-form robot that mimics human behavior or behaves in ways similar to humans. I’m not talking about androids.

What I envision are machines that are designed to move gingerly through thick growth, performing simple operations like planting seeds and pulling weeds, while creating a minimum of disturbance. They might be supported on long, spider-like legs that only ever put weight on particular, gravel-covered spots, or they might be suspended from overhead rails or cables, but they would be engaged in paying detailed attention to what was happening on the ground, including the presence of animals as well as the slow-paced growth of plants, intervening only occasionally as compared with the amount of information they would be processing, although to the casual observer it might look like they were awfully busy.

These machines would have only a modest rate of energy consumption, and might easily get what they need from solar panels, with any excess going to charge batteries that would allow them to continue to operate into the night, and at least keep watch 24/7.

I can only envision such machines, of course, because they don’t yet exist.

Reposted from Cultibotics.

From an ecological point of view, robots can help replace a factory-like system of crop production, that slowly turns fields to deserts, with a more nature-compatible, nature-reliant system that can restore fertility.

From a dietary standpoint, the use of robots can [enable a vast] increase in the variety of crops produced on any piece of land, and help make [healthier] alternatives to a heavily grain-based diet more affordable.

From the perspective of the consumption of nonrenewable resources, like petroleum, assuming they aren’t just doing the same old things, robots can get by on a fraction of the energy spent on tillage, and because they would do their work slowly and continuously instead of all at once, could get much or most of their energy from renewable sources (sun and wind).

There are other reasons, as well as much more to be said about these. I’ll get to it in time.

Reposted from Cultibotics.

Well, so be it, for now. There is good work being done, and as the variety of off-the-shelf robotic parts continues to increase, the power/price of computer equipment continues to rise steeply, and the economics of current agricultural practice continues to degenerate with rising costs for fuel and other inputs, the scene is gradually being set for a profound transformation in the way we use land to produce food, fiber, and plant materials for the production of synthetic fuels.

There is a danger that we won’t get serious about that transformation until the breakdown of current practice is far enough along to cause serious disruption and pain. There is also danger that robotics will first be used to put off real change as long as possible, by simply displacing what few humans remain in an otherwise essentially unchanged system.

On the other hand, there is plenty of opportunity to go around, especially for those who get in on the ground floor and develop the necessary technology to augment generic robotics and create energy-efficient machines designed to perform detailed management of productive land in ways that enhance fertility and repair ecological damage.

The demand for agricultural production isn’t going to go away, but we can dramatically change and radically improve the way we go about meeting it, with a far greater return on the R&D dollar than, for example, missile-bearing robotic helicopters.

Reposted from Cultibotics.

Don’t let the title scare you off. There’s more here than might at first be apparent.

There’s much to suggest that the original condition of human beings was nomadic – maybe over a fairly limited range, but seldom entirely sedentary.

First farming, then cities, with the investment of time and effort they embody, made moving about with the seasons less appealing. But in modern times we seem to have replaced both nomadism and attachment to place with a sort of serial squatting, staying a few months in one place and a few years in another, but generally without the deep connections that come from spending one’s whole life in the same place and among the same people.

But we aren’t likely to ween ourselves from dependence on stationary infrastructure anytime soon, not even with the recently booming popularity of mobile devices, and in that realization is the germ of the thought I hope to develop here: people want the benefits of infrastructure, without being bogged down by it.

And, increasingly so as time goes on, infrastructure means not just brick and mortar, but something active.

Asimov described a planet sparsely inhabited by people who were heavily dependent on technology, who each had substantial land holdings which were cared for by robots.

All very scifi, but the connection between machine and place rings true.

Most of us cringe at the idea of a life devoted to looking after a plot of ground, yet the ground needs looking after, particularly in and near cities. How strange is it really to propose that machines will inherit this task, probably sooner rather than later?

Here are some examples of familiar technological intrusions on the landscape: roads, bridges, dams, irrigation ditches, water and sewer lines, storm drains, tilled fields, fences, hedgerows, terraces, power lines, rail beds, mines, harvested forests (and those replanted with less than their original diversity), not to mention buildings.

We’ve never much hesitated to impose our notions on the world around us, but until recently we did so personally, typically using the biggest time-saving levers we could find. Only we (and a few domesticated allies, like sheepdogs) could provide the element of attentive activity, for lack of which one built environment after another has reverted back to a wild state, generally an impoverished one. We’re good at dreaming up grand designs, and we like having things neat and tidy, but we tire of maintenance, and are more than a little stingy about paying for it.

In suggesting that we embrace a robotic approach to land [management], I’m also implying an eventual end to capricious land use, with grand new schemes equating to changes in programming that would result in gradual modifications of the landscape, rather than sudden, gaping scars that take decades to heal.

Reposted from Cultibotics.

This developerWorks article discusses open source software for modeling and testing robotic designs in software.

(Found on AI Buzz. [2012Oct13: now a Japanese language website])

Reposted from Cultibotics.

Granted that they’re using chemical film in the camera itself, but that film is scanned as soon as it’s processed, and all subsequent image manipulation is done digitally. The file resulting from a full-resolution scan of one such film is 24 GB, which does push the limits of current technology a bit. [2012Oct13: larger thumb drives now available]

While really not at all the same as what a cultibot would do in cataloging all of the plants within the area it tended, this does suggest that the level of complexity involved is within the reach of either current technology or what’s just around the corner. Maybe the database for a five-acre plot would occupy a few terabytes, but even that isn’t unmanageable. [2012Oct13: easily within desktop RAID capacities today]

Reposted from Cultibotics.

So when I saw a pointer to a Spiegel article titled Bringing Robot Transportation to Europe on Slashdot, I had to see what it was about. [2012-Oct-11: This article is no longer available at the address that was originally linked.]

As it turns out, that article describes a type of system which the Innovative Transportation Technologies website has been tracking for a decade.

Reposted from Cultibotics.

IEEE Robotics and Automation Society’s Technical Committee on Service Robots website contains a page dedicated to Agriculture & Harvesting Robots, with many links to related activities around the world. [2012-Oct-11: The RAS Technical Committee on Service Robots no longer maintains a website.]

Arrick Robotics’s website has a long list of robotics clubs. One of those, the San Francisco Robotics Society of America has a particularly interesting website. [2012-Oct-11: Still true!]

Robots.net, Robot Gossip, and GoRobotics.net are robotics news blogs.

Artificial Intelligence News, just what it says it is. [2012-Oct-13: Now it points to a Japanese language website.]

I only plan to cover developments which are at least plausibly related to the application of robotics to crop cultivation, except perhaps as a bit of news is so compelling that I can’t help but mention it.

Reposted from Cultibotics.

There’s a more general sense of “cultivation” that simply refers to raising plants, and which applies as much to the practices of nomadic tribes, involving no more tillage than poking seed holes in the ground with a sharp stick, as it does to agriculture as it is commonly practiced today.

It’s that second, more general sense of the word that’s intended here, as the potential advantages of using robotics in horticulture and agriculture stem largely from making it possible to dispense with the other sense of the word, tillage.

That’s not to say that a cultibot wouldn’t perform soil manipulation, rather what’s expected is that it will resemble what a gardener performs with a hoe, trowel, and (occasionally) shovel, instead of what a farmer performs with a plow, and that the total amount of energy involved in performing it will be a small fraction of what the current practice of farming consumes – and the rate of energy consumption even lower, since autonomous operation will allow it to be distributed over more time, perhaps even 24/7. (This combination of lower energy requirements and more time should make solar panels a practical power source.)

There’s yet another sense of the word, as it applies reflexively or to human relationships, as in the cultivation of patience or friendship, which should at least inform how the vision of cultibotics is understood. In its fully realized state, a cultibot would not only raise plants and produce food, but it would tend the land in all its aspects, specifically including as it also serves as habitat for wild species, both plants and animals. This could be seen as cultivating a field’s participation in the larger environment, making a little room among the crops for other life.

Reposted from Cultibotics.

Potential advantages include reduction or elimination of the need for petroleum-based fuels, fertilizer, and pesticides, an increase in the variety and value per acre of crops produced, a huge improvement in the sustainability of agriculture, and a revitalization of rural society through a more interesting, varied environment and the creation of technical jobs (maintenance, etc.). Detailed, automated land management could also help rescue endangered plant species from the brink of extinction.

I’ll be gradually filling out this vision and substantiating each of the points above, while at the same time mentioning any related developments I might learn about and accumulating a list of related efforts (academic, commercial, etc.).

Reposted from Cultibotics.