Why Agriculture Can Never Be Sustainable, and a Permacultural Solution present by Toby Hemenway

This video was recorded two years ago, but only recently posted to YouTube. I think it's amazingly good!

Biological Agriculture for Roboticists, Part 5

To be most useful, agricultural robots need not only to be able to distinguish plants from a background of soil and decaying plant matter, but to be able to distinguish them from each other, and to quickly model their branching structures, at least approximately, if only so they can locate the main stem and the point at which it emerges from the soil. They also need to be able to recognize plants that don't belong to any of the types they've already learned to identify as being something new.

This is a tall order, and I'll get into some specifics on how it might be accomplished a bit further on, but first why would robots need to be able to recognize plants as being something they haven't seen before; isn't it enough to be able to tell whether they've been planted intentionally, crop or not?

In Part 4 of this series, I provisionally claimed that, in a recently tilled field, which has not yet been planted to the next crop, any green growing thing can be presumed to be a weed. While that's usually the case there are exceptions.

Even in a monoculture scenario with routine tillage, where you don't really expect to find anything other than the crop that the farmer has planted and weeds in the field, seed may be brought in from elsewhere, blown on the wind, in bird droppings, or in the stools or clinging to the fur of some wide-ranging mammal. Generally these also might be considered weeds, but occasionally they will be rare and endangered species themselves, or vital to the survival of rare and endangered animals (milkweed for monarch butterflies), and should therefore be allowed to grow and mature, even at the expense of a small percentage of crop production and some inconvenience. (Farmers should be compensated for allowing this to happen, and robotic equipment can help document that they have done so.)

In a poly/permaculture scenario, native plants that aren't poisonous to livestock or wildlife, and which don't compete aggressively with crops, are usually welcome, because they increase the diversity of the flora, supporting a more diverse fauna, which is more likely to include beneficial species, all of which implies a more stable environment, less prone to overwhelming infestations of all sorts.

Plants look different under different lighting conditions — dawn, mid-morning, noon, mid-afternoon, dusk, and under clear sky versus clouds versus overcast conditions — and different in shade than when standing alone on otherwise clear ground. Beyond that, plants look very different as seedlings than they do after a few weeks of growth, different yet when they've gone to flower, and different once again when mature, and for deciduous perennials still more different in their winter or dry season dormancy. Without having seen them in all of these stages and conditions, even a human gardener might mistake one crop plant for another, or for a weed, and based upon that select an inappropriate action. Recognizing continuity between stages and across diverse conditions is even more challenging for a machine.

For all of these reasons, once the technology is up to making such differentiations quickly enough that it is no longer the limiting factor in machine performance, the default needs to be that when confronted with something unfamiliar do nothing other than keep track of it, and send a notification up the escalation chain. Now back to the question of how, which is about sensory modes and sensory processing. What information about an environment composed of crops, a smattering of native plants, and weeds, on a background of soil and decaying plant matter, can a machine usefully collect and process?

Among the most obvious and most valuable is location. To a very close approximation, plants stay where they're planted, so if today you find a plant in the same location as you found one yesterday, there's a high probability that it's the same plant, just a day older. (It's true that some plants send up new shoots from their root systems, remote from the original stem, but that belongs to a discussion of modeling things that aren't directly sensible, or, in that example, requires something like ground-penetrating radar.) Generally speaking for plants, over a short interval, location is synonymous with identity. GPS by itself is inadequate to establish location with sufficient precision to be used in this manner, so it must be supplemented with other methods, such as fixed markers, odometry, and maps created on previous passes over the same ground. More precise local positioning systems could also prove very helpful.

Another obvious collection of modalities center around imagery based on sensing reflected electromagnetic energy, including everything from microwaves through infrared and visible light to ultraviolet, as snapshots and over time (video), and using ambient or active illumination, or a combination of the two. (Introduction to RADAR Remote Sensing for Vegetation Mapping and Monitoring) Color video camera modules have become so inexpensive that using an array of them has become a reasonable proposition, and modules containing two or more lens/sensor systems are becoming widely available. Cameras which are sensitive to ultraviolet, near-infrared (wavelengths just longer than visible light), and far-infrared (thermal radiation) are also becoming common and dropping in price. Even phased array radar is being modularized and should be reasonable to include in mobile machines within a very few years.

Other sensory modes that are either already in common use, or may soon be so, include sound, both passive (hearing) and active (sonar), pressure/strain (touch-bars, whiskers, and controlling manipulator force), simple gas measurement (H2O, CO2, CH4) and volatile organic compound detection (smell, useful in distinguishing seedlings). I'll get back to the use of sound in a future installment, in the context of fauna management and pest control.

The stickier problem is how to transform all the data produced by all available sensors into something useful. This can be somewhat simplified by the inclusion of preprocessing circuitry in sensor modules, so that, for example, a camera module serves processed imagery instead of a raw data stream, but that still leaves the challenge of sensor fusion, weaving all of the data from all of the various sensors together to create an integrated model of the machine's environment and position within it, both reflecting physical reality and supporting decisions about what to do next, quickly enough to be useful. Again, research is ongoing.

Previous installments:

• Biological Agriculture for Roboticists, Part 1
• Biological Agriculture for Roboticists, Part 2
• Biological Agriculture for Roboticists, Part 3
• Biological Agriculture for Roboticists, Part 4

Biological Agriculture for Roboticists, Part 4

So how can robotics contribute to agriculture, or, more generally, to land management? Let's start with a relatively simple example, where the robot need not concern itself with differentiating between crops and weeds, and the only required manipulations are of nonliving materials. I'm talking about erosion control. In the video linked below there is no mention of robots, but the speaker does describe one of the techniques he employs — close placement of stones such that they aren't vulnerable to being washed away in the next flood — as time consuming and tedious. This is almost certainly a task that could be automated.

One step up from this is weed elimination in a recently tilled field which has not yet been seeded for the next crop, sometimes referred to as 'fallow' although that word is also used to mean a field that is simply being left alone for a time and has not been tilled. In that recently tilled field, any green growing thing can be assumed to be a weed — well, not exactly, but for the present purpose yes; I'll get back to this in a future installment — so all the robot has to be able to do is differentiate between the color of tilled soil and the color of any green growing thing within a geofence. Spot application of herbicide directly onto the plant, scaled to the size of the plant (less on smaller plants), is a huge improvement over area-wide application of the same herbicide, because it results in much less being used, and there are already machines available that do this.

Still using herbicides, further improvement is possible if the robot can model the plant in some detail, instead of only detecting a green blob, and determine where various parts of the plant are in space and relative to each other, how they are attached. Given such a model, more precise application of still smaller amounts of herbicide becomes possible, dropwise onto the point or points where cell division takes place (the meristem or meristems) or, in the case of an established plant, by injection into the main stem, just above where it emerges from the soil. Drop application can be tricky in a breeze, with the plant moving around, but if that model includes the structure of the plant and how it moves in response to air currents — something that is extractable from video (see below) — then those motions can be predicted and compensated for in the positioning of the drop dispenser. Other methods, not involving herbicides, are also possible, and made a great deal easier by this sort of modeling.

Pastures are a bit more complicated than recently tilled fields, since the contrast between weeds and everything else is more subtle, but help is available. Herbivores can be picky about what they eat and don't eat, and what they don't eat can produce seed or spread vegetatively, reducing the value of the pasture. In combination with holistic management of the animals themselves, keeping unpalatable plant species in check can help protect and improve pasture land. This is a task that can be performed by robots, at its simplest by clipping off at ground level anything left standing more than a few centimeters high immediately after a paddock has been grazed, after the animals have already decided for themselves what's good to eat and what isn't. This approach won't control low-growing plants, like goatheads (photo courtesy of Forest & Kim Starr), but it will control thistles and other erect plants. (Again, more in a later installment.)

Weed control becomes more challenging when the machine must be able to distinguish between crops and weeds, and research into accomplishing this is ongoing, but given that perceptual ability, the ways in which a robot might deal with a weed are numerous, and generally fall into three categories, control of seedlings, control of established plants that can easily be uprooted, and control of established plants that break off if you attempt to uproot them, leaving the root behind to produce a new shoot.

Seedlings are difficult to distinguish, but easy to kill. Most are easily uprooted, but moving a mechanism into position to do the uprooting takes time, especially if there are many weed seedlings to be dealt with in this manner. Using a high-pressure water jet to expose the root or sever the stem should take less time, since no movement reversal (pulling) is required. Using projectiles (ice, dry compressed compost, ...) might accomplish much the same thing from a slightly greater distance, given accurate targeting. With even more precise targeting, a laser might heat just the meristem enough to stop a seedling's development, from an even greater distance, requiring even smaller movements for retargeting, enabling faster operation. Laser heating might be combined with LIDAR sensing as a precisely timed high-energy pulse. A downside for both projectile and laser methods is that they require a clear path from a remote position to the target, something that becomes more problematic as the growing season progresses and leaf canopies become more dense, but seedlings that emerge later in the season are less of a concern, both because they will develop more slowly due to the shade created by established plants and because, by the time they have developed enough to represent significant competition for resources, annual crops will already be maturing or have already been harvested.

Plants with established root systems can sometimes also be uprooted, and any established plant can be clipped off mechanically near the soil surface, or cut off with a jet of high-pressure water. All of these techniques require positioning a mechanism at or near the base of the plant, and uprooting also requires support sufficient not only for the weight of that mechanism but also to offset the force required to accomplish uprooting, which can be considerable. Clipping or cutting a plant off near the soil line may not kill it, but it will set back its growth, and doing so repeatedly can eventually exhaust the resources it draws upon to regenerate above-ground growth, provided that the root system isn't being fed by foliage elsewhere. Machines relying upon such methods should be programmed to revisit those locations periodically, checking for regrowth. Plants with very tenacious root systems may require more aggressive treatment, which could mean herbicides but might also mean coring (cutting a deep cylindrical plug from the soil around the plant's stem) or steam injection, to cut the node from which that stem emerged off from the rest of the root system or locally kill the root system. All of this is a little like the botanical equivalent of Whac-A-Mole, which means tedium, something robots excel at coping with.

Next I'll go into perceptual systems (sensing and sensory processing) and plant differentiation in greater depth.

Previous installments:

• Biological Agriculture for Roboticists, Part 1
• Biological Agriculture for Roboticists, Part 2
• Biological Agriculture for Roboticists, Part 3

Robots Podcast #208: James Underwood

Robots Podcast #208 is an interview with James Underwood, who is a senior research fellow at the Australian Centre for Field Robotics, University of Sydney. He “is an expert in the area of perception systems for field robotics – the study of how outdoor robots working in complex, unstructured environments can make sense of their world using science and technology in multi-modal sensing, data fusion and mapping.”

In that interview, Dr. Underwood goes far beyond discussing the featured Ladybird project to provide us with an excellent overview of where agricultural robotics is headed over the next few years.

Be sure not to miss this podcast!

Biological Agriculture for Roboticists, Part 3

Now you might wonder why I'm referring specifically to 'Biological' agriculture, rather than just refer to agriculture in general; allow me to explain.

The loss of soil through erosion, and the loss of fertility and the capacity for water retention in what remains are both reaching crisis proportions, with the result that it will become increasingly difficult to maintain yields on much of the land currently in production. Compounding this is the prospect of a world population that may be brushing up against ten billion by the middle of this century, the spread of urban areas onto land previously used to produce food, and climatic changes that dictate changes in what crop is grown where.

Moreover, while the number of people afflicted by starvation is actually falling, the percentage of people affected by malnutrition is on the increase. Over the last few decades, we have been catching up with the demand for calories, even for protein, but falling behind in terms of a well-rounded diet. If the goal is to feed everyone to a high standard, then we are not on track to achieve it, even without the further loss of production capacity.

There are countertrends, of course. Consumer cooperatives, urban farming, and direct connections between market farmers located near cities and consumers within them all work to narrow the gap between what is available and affordable through conventional channels and that well-rounded diet. On the downside, local production is even more seasonal than are conventional channels, which regularly bring in goods harvested unripe to help them survive long-distance transport, selling these at a premium in the off-season. It remains to be seen how much of that gap can be spanned for how many how soon. There are many working to push the envelope as far as it can be pushed, but habit and an avalanche of advertising stand in the way of rapid change.

One specific countertrend where robots are already making an inroad is in the production of leafy greens under artificial light in urban factory settings. Because these are controlled environments with repetitious geometries, it is relatively easy to make machines operate within them. No doubt this trend will continue to make headway, branching out into other types of produce at it proves practical to do so, but, at least for the near term, it will probably be limited to crops that are shade tolerant, not requiring intense light, and even the seed for these will continue to be produced outside.

If these trends continue to gain momentum, they may take some of the pressure off rural food production, even with a growing population. Nevertheless, we will continue to need that rural production for a long time to come, for the foreseeable future, so we'd better be thinking about how we can protect and improve the productivity of the land we depend upon.

This is where both biological agriculture and robots come in. In recent times, while what constituted conventional agriculture has been growing in scale and increasing its use of industrially produced fertilizers, pesticides, and herbicides, the practice of biological agriculture has mostly happened at the scale of gardens worked by hand, and, for the people working these gardens, this is quite often simply a matter of using traditional methods passed down through generations, which aren't necessarily optimal, but which are generally less destructive than the methods that have become conventional in the developed world. Unfortunately, such traditional methods only scale with the number of hands available to perform them, and the trend has been for people to leave this way of life for the cities – a new wave of emigration with each new war or environmental catastrophe – with the land they had been tending falling into the hands of larger scale farmers using more conventional methods. (Take a minute to consider for whom this migration off the land might be convenient.)

Using conventional equipment, the methods of biological agriculture don't scale without compromise. Where you might pull weeds by hand and avoid tillage in a small garden, as the size of the farm goes up the temptation to use wholesale methods, tillage or herbicides or both, can become overwhelming. Tillage exposes the soil to erosion, and herbicides can mess with soil microbiota, leach into ground water, are washed away with runoff, and remain present in trace amounts in the food produced, also true for pesticides and fungicides. Granted that it's not really fair to lump all such products together as being destructive – some are far more so than others – but on balance it's better to avoid their use if a practical alternative exists, and, if they must be used, to use them as sparingly as possible, through precise application.

What's needed is equipment capable of using methods which are functionally equivalent or superior to what a gardener doing the work by hand might use, and can do so without a human operator, so the number of such machines in operation is not limited to the number of human operators available. Perhaps surprisingly, the second of these requirements frees equipment manufacturers to opt for smaller rather than larger designs, both reducing the amount of damage one might do if it malfunctions and enabling economies of scale not available to equipment produced only in the thousands of units per year.

What may also be surprising is that, while equipment designed to work thousands of acres doesn't scale down for use in gardens, equipment designed for use in gardens, if it can be trusted to operate autonomously, will quite happily scale up to thousands of acres, bringing with it a transformation of the methods in use.

In the absence of ready examples to point to, it's hard to know how the economics of this transformation might work out. While the smaller devices would each be far less expensive than, for example, a modern tractor, many more of them would be needed, so equipment cost might actually be higher. On the other hand, if they are doing the mechanical equivalent of weeding by hand, and not consuming diesel fuel to power pulling implements through the soil, the cost of other inputs will be reduced. Also, their use may enable the cultivation of higher value crops that can't easily be produced by bulk methods.

Improving technology and techniques will undoubtedly, at some point, tip the balance in favor of using smaller, autonomous machinery, so that farms which do so are more profitable than those that do not, and once that happens further development will be driven by a rapidly growing market.

The smallholders alluded to above will be among the biggest beneficiaries of this trend. Much of the technology developed for conventional agriculture has only made their lives harder, by driving down the prices of commodity crops like maize and wheat. But the development of small, autonomous equipment, well suited to small plots and crops that aren't so easily produced using bulk methods, would make it possible for them to use more intensive management of the land they have, and spend less time doing it.

Another class of beneficiary is plant breeders, the people working to develop crop varieties that are better at competing with weeds, are more resistant to pests, are more hardy under a wider range of environmental conditions, produce higher yields in poorer soil with less fertilizer, and, as if all that weren't enough, are also more nutritious. This is tedious work*, and machines capable of performing detailed field operations autonomously would dramatically improve its effectiveness, in particular because land in active production and the machines tending that land could all be enlisted in the effort, by simply including some additional code in the programming of those machines. That increased effectiveness could help insure the availability of nonproprietary seeds which a farmer can save from one crop to plant the next without the next generation being far less productive than the original seed, and without being sued for doing so.

Next I'll get into some specifics of just how robots might perform such detailed field operations.

Biological Agriculture for Roboticists, Part 2

Not every plant we cultivate for food grows from seed in soil. Some don't grow from seed, either naturally or by our choice, because we want to perpetuate the characteristics of a particularly useful genome.

Vegetative Reproduction

Any plant reproduction that doesn't involve the reshuffling of chromosomes which results from sexual reproduction. Wikipedia has an excellent article on this topic.

Plant Propagation

This is a more general term that includes both reproduction from seed and vegetative reproduction. Again, Wikipedia has this covered.

There are also some plants that naturally don't grow in soil, but here we're mostly interested in cultivation methods not involving soil, except perhaps as an inert granular medium used only for mechanical support.

Aquaculture

Generally thought of as a way of producing fish, crustaceans (shrimp, lobsters, crayfish) or gastropods (snails, clams), aquaculture can also be a way of producing food those animals eat, or plant matter for other uses. Everything from suspended phytoplankton to vascular plants anchored on the bottom or floating on the surface can contribute to the biomass available to/from the system. Besides being productive in itself, aquaculture combines well with other practices, because, whether due to the presence of microorganisms or the waste products of animals, the water makes good fertilizer.

Hydroponics & Aeroponics

These are two variations on the theme of growing plants in a nutrient solution without soil, although, in the case of hydroponics, an inert granular medium may be used for mechanical support. Aeroponics is distinguished by the use of mist to keep the roots moist, rather than fluid water, obviating the need for supporting structures sufficiently strong to hold up the mass involved in using large amounts of water, and also avoiding the possibility of anaerobic conditions developing.

Aquaponics

Aquaponics is the combination of aquaculture and either hydroponics or aeroponics, where water from aquaculture is used as a fertilizer, and perhaps also plants produced using that water are fed to animals in the aquaculture environment. This cycling of nutrients can result in a highly productive systems which make good use of both space and available light energy.

While aquaponic systems are very good at producing quick-growing leafy greens, like lettuce, and protein from aquatic animals, the cost of infrastructure makes the use of hydroponics or even aeroponics prohibitively expensive for the production of many crops, and for others, like potatoes and peanuts, they are simply unsuitable because the part of the plant we are interested in harvesting grows below the soil surface, or, like wheat, the plants need a dry environment for final ripening and are susceptible to fungal damage if the humidity is too high too late in the growing season. So, we'll be needing soil for awhile yet.

Either way, whether using soil or not, aquaculture can be an integral part of the system.

Also either way, whether using soil or not, whether outdoors, or in polytunnels or permanent greenhouses, or in racks under light from LEDs, or even growing mushrooms in the dark, there's a place for robots, lots of robots, maybe even billions of robots.

Biological Agriculture for Roboticists, Part 1

This is the first post in what I expect will become a series. As such, it makes no attempt to be comprehensive, but rather is intended to be a gentle introduction to the subject, which I have chosen to call "biological agriculture" for an audience composed of roboticists and robotics enthusiasts. This choice of names is somewhat arbitrary, but I think it will serve well enough, the idea being that methods based in biology and the manipulation of organisms should be used in preference to methods based in chemistry, even biochemistry. By this I do not mean that there is no place for chemistry, far from it, but that, in the sense/think/act paradigm common in robotics, chemistry properly belongs primarily to sensory input.

I also expect to do a series on Robotics for Gardeners and Farmers, as time and clarity allow.

Let's begin with a few definitions.

Cultivation

Cultivation has two primary meanings, one of which is specific to a class of tillage operations in which the soil is disturbed, usually lightly, to kill or suppress the development of weeds. The other meaning of cultivation is more general, roughly equivalent to the term husbandry, and may or may not include tillage as a practice. This more general meaning is what I had in mind in naming my blog Cultibotics. Cultivation without tillage is frequently referred to as a 'no-till' regime.

Soil

Soil is more than dirt. It is a complex ecosystem generally including a mineral component, but even that mineral component isn't static. The clay in soil is constantly in flux, trapping nutrients as it forms and releasing them again as it dissolves. It also contributes to the ability of soil to absorb and retain water. Decomposing organic matter in soil is collectively referred to as humus, which also contributes to water retention, as well as providing both nutrients for plants and food for the many living organisms found below the surface. Besides a bewildering array of bacteria and fungi, these also include worms, insects, and higher animals like moles and mice. Some microorganisms live in symbiosis with the roots of plants, receiving energy from them and providing a valuable service in return, for example nitrogen fixation. Living soil is also partly gaseous, with a continuous network of tiny pockets and channels throughout. In the absence of tillage or other disturbance, soil develops structure driven by the many simultaneous processes happening within it.

Fertilizer

Anything that provides substances needed by plants for growth and maturation can be considered a fertilizer, from the dung of passing herds of ungulates to composted feedlot waste, to slurry from urban sewage, to ground minerals, to industrially produced ammonium nitrate. Fertilizers are usually applied for the purpose of increasing yields, but may also be used to improve the nutritional quality and/or marketability of crops produced.

Herbicide

You may have heard of Glyphosate. It is one prominent example of herbicides, compounds marketed for the purpose of killing or suppressing the development of weeds that compete with crops, and has become a hot topic of debate in part because crop varieties which are relatively tolerant of this substance have been developed to broaden its applicability to crops that would otherwise be damaged or killed by it as much as the weeds it's meant to control. This is because it is a 'broad-spectrum' herbicide. Some herbicides are more specific, killing or stunting the growth of problematic species more selectively.

Pesticide

Generally the term pesticide is applied to substances which kill or suppress the development of multicellular animals, usually arthropods (insects, spiders, crustaceans) but commonly also tiny worms called nematodes and gastropods (slugs). Some of these substances are produced by and refined from organisms, or synthesized based on naturally produced compounds, while others are entirely synthetic. Depending on the metabolic pathway they target, pesticides may also be highly toxic to humans, and the use of those that are has already been prohibited or is tightly regulated, with limits established for residues in food sold for human consumption.

Fungicide

Fungicides are much like pesticides, except that they target fungi, such as ergot, which infects cereal crops like wheat and rye.

Biological Control

A prime example of biological control is one used by many gardeners, the release of ladybird beetles (ladybugs) to control aphids. Others include distribution of praying mantis egg cases and encouragement of parasitic wasps, which lay their eggs in or on the larvae of a variety of insects, such as the tomato hornworm. Biological controls tend to be more specific than synthesized products and are far less likely to present any danger to humans, but they also tend to be more management intensive, meaning that, for a given effect, they tend to require more time and attention from the gardener or farmer.

Polyculture

Plants are seldom found in uniform, single-species stands in nature, there is nearly always a mixture of species, forming plant communities, which in turn constitute the most visible aspect of ecosystems. Polyculture attempts to mimic this by combining plants that play well together into mixed plantings, simultaneously, sequentially, or both. One popular approach which arguably qualifies as polyculture is planting different crops in alternating rows in the same field. This practice may help keep populations of beneficial insects from crashing, by spreading out the season during which resources are available to them.

Crop Rotation

A compromise solution which achieves some of the benefits of polyculture without the complication of combining multiple types of plant together in the same field at the same time is crop rotation. Rotations can often be expressed as a simple, repeating sequence of crops, such as clover followed by maize followed by wheat.

Permaculture

Think vineyards and orchards, although some herbaceous crops are also perennial. Even though vines and trees eventually die, they live for years and may be replaced by young plants of the same type, perpetuating the continuity of the ecosystem which develops around them. Permacultures may also be polycultures, and may include annual crops in the mix, but the presence of perennials constrains the use of highly disruptive techniques, like plowing, to the space between them, discouraging the use of such techniques at all.

That's enough for a decent beginning. Stay tuned for Part 2, and for the series on Robotics for Gardeners and Farmers.