Pollination by bees and other insects is one of the natural and essential biological processes for about three-quarters of global crop species. However, pesticides, land clearing and changes in climate have been associated with declines in the number of many of these living creatures.
Over the course of many years, ways of pollinating plants have abounded, with limited success. For instance, hand pollination (with a cotton swab or a small brush), while effective, is labour intensive. And machine pollination with pollen blowers, dusters and spray dispensers can be expensive.
Almost two years ago, a group of researchers from the Japan Advanced Institute of Science and Technology (JAIST) published a study in iScience suggesting that soap bubbles could be used as a low-tech method of artificial pollination. Their experiments showed that chemically functionalized soap bubbles exhibit effective and convenient delivery of pollen grains to targeted flowers thanks to their stickiness, softness, high flexibility and enhancement of pollen activity.
JAIST researchers Xi Yang and Eijiro Miyako designed and prepared chemically functionalized soap bubbles for the artificial pollination of flowers using various bubble-making devices. Their methodology included the use of an unmanned aerial vehicle equipped with a soap bubble maker that was autonomously controlled to pollinate flowers.
JAIST researchers posit that soap bubbles could be used as a low-tech method of artificial pollination. Photo: E. Miyako, JAIST
In their experiments, the researchers noted high controllability of the directional flying of soap bubbles by bubble-making devices is useful for simply shooting soap bubbles directly onto the target flowers so as to systematically reduce the workload of thinning out superfluous fruits. “Moreover, scattering of pollen grains can be physically restrained as they are tightly confined to the liquid membrane of the soap bubbles,” they said.
In their experiments, the researchers successfully pollinated the flowers of P. pyrifolia var. culta in an orchard by soap bubbles using a bubble gun, which consequently formed young pear fruits. They also integrated soap bubbles within a drone for fully automatic pollination of L. japonicum flowers.
“These chemically functionalized soap bubbles demonstrated interesting properties, such as the delivery of pollen particles, enhancement of pollen activity, excellent mechanical stability, suitable size and geometry of flower pistils, softness and high flexibility to prevent damage to flowers, adhesiveness on flowers for effective pollination, and simple ejection ability on a large number of soap bubbles from the devices,” they noted.
The JAIST study was the first to explore the unique properties of soap bubbles as a material used for the artificial pollination of flowers using different types of bubble-making tools and an autonomous controllable drone.
“We expect our multidisciplinary approach combining soap bubbles and drone technology to lead to innovative developments in the field of agricultural engineering,” the researchers posited.
Meanwhile, an artificial pollination company based in Israel, Edete Precision Technologies for Agriculture, has developed an end-to-end artificial pollination service comprised of two steps that mirrors the action of honeybees – collecting and distributing pollen.
Edete’s mechanical pollinator units can operate day and night. Photo: Edete Precision Technologies for Agriculture
The first step is to harvest the pollen. Edete mechanically harvests flowers, separates the pollen from the flower, then stores the pollen.
The second step is pollen distribution. Specially equipped vehicles drive down orchard rows gently blowing out the pollen, which is given an electrostatic charge to keep the individual grains from sticking together. The vehicles use lidar sensors for precision, staying within 10 centimetres of the trees’ contours. Edete stated its’ mechanical pollinator units can operate day and night, rapidly and thoroughly covering any open flower in its range, and are not limited by daylight or low temperatures conditions.
Robotic crop pollination, too, is proving advantageous as a form of artificial pollination, so much so that developing and designing of intelligent functional robots is an ever-growing emerging technology.
Inspired by the biology of a bee, researchers at Harvard University’s Wyss Institute have developed RoboBees, autonomous flying microrobots that could act as crop pollinators. The RoboBee measures about half the size of a paper clip, weighs less than one-tenth of a gram, and flies using “artificial muscles” compromised of materials that contract when a voltage is applied. The RoboBee development is broadly divided into three main components: the body, brain and colony. Body development consists of constructing robotic insects able to fly on their own with the help of a compact and seamlessly integrated power source; brain development is concerned with “smart” sensors and control electronics that mimic the eyes and antennae of a bee, and can sense and respond dynamically to the environment; the colony’s focus is about coordinating the behaviour of many independent robots so they act as an effective unit.
RoboBee, an autonomous flying microrobot that could act as a crop pollinator.
Photo: Wyss Institute at Harvard University
Just last month, RoboBee hit another milestone: precision control over its heading and lateral movement, making it much more adept at maneuvering. As a result, RoboBee can hover and pivot better in midair, more similarly to its biological inspiration, the bumblebee.
Other robotic pollination projects abound around the world. The head of robotics at Israel’s Ben-Gurion University of the Negev, Yael Edan, spoke a few years ago about her lab group’s research in developing a mini-drone that replaces the need for bees in pollination. In 2018, Walmart applied for a patent for drone pollination, perhaps a sign the company hopes to venture into agriculture and gain more control over its food supply chain. And a group of scientists at Delft University of Technology in the Netherlands believe they will be able to create swarms of bee-like drones to pollinate plants when the real-life insects have died away. The robo-bees, called DelFly, can hover on the spot, fly in any direction and even flip 360 degrees around pitch or roll axes. Because the robots’ wings are made of a lightweight film made of mylar, the material used in space blankets, it is safe for people to work around them. These new drones, which can travel at up to 15 mph, are also more efficient in their flight than those with helicopter-style blades, meaning their batteries can last longer. They can be fitted with spatial sensors so that they autonomously fly from plant to plant, avoiding each other and other obstacles as they go.
Computer-generated image of StickBug, a six-armed robot to assist humans in greenhouse environments by pollinating various crops, which is being developed by WVU researchers.
Photo: WVU Robotics
More recently, Washington State University researchers are leading a team that’s exploring the use of robotic pollinators to assist fruit farmers. In 2020, the project received a three-year grant totaling nearly USD$1 million from the U.S. Department of Agriculture through the Washington State Department of Agriculture. The team is building on and adapting existing machine learning and robotics technology, including tools being developed to mechanically thin plants or pick fruit such as apples.
Four years ago, researchers at West Virginia University (WVU) developed a robotic pollinator, the BrambleBee, that operates similarly to a self-driving car. If BrambleBee determines a flower is ready for pollination, it will use a small 3-D printed brush of flexible polyurethane bristles on the end of its arm to gently stroke the blossom to assist with the transfer of pollen from the anthers to the pistils. The robot will even remember what flowers it already hit, so it can make multiple runs as the plants mature in the greenhouse.
In 2021, WVU researchers created a new robotic pollinator – StickBug – a six-armed robot. The project was led by Yu Gu, associate professor in WVU’s department of mechanical and aerospace engineering, and joined on the project by Jason Gross, associate professor and associate chair for research, mechanical and aerospace engineering, and Nicole Waterland, associate professor of horticulture and director of controlled environments.
“It (StickBug) maps out the environment and once the robot has a general idea of the environment, it will build up a more detailed mapping of the plants and knows where the flowers are and which flower needs to be pollinated,” Gu said. “It will make a plan on what to do. Then, it will get close to each of the plants, start swinging its six arms and start pollinating.”
According to Gu, the six arms are mainly for improving the efficiency and effectiveness of the robot. For example, some flowers could be in hard-to-reach places and the robot may need to use two arms. One arm for grabbing the branch, and the other arm to pollinate the flower.
The long-term goals for this robot are to care for individual crops efficiently, improve food security during insect declines, support indoor agriculture and provide services beyond what insects can do such as collecting data on the crops.
With a very high percentage of crops around the world that rely on pollination, the aid of artificial and robotic pollinators is certainly conceivable. New AG International be keeping an eye on the technology as it evolves. ●
Once the robot has a general idea of the environment, it will build up a more detailed mapping of the plants and knows where the flowers are and which flower needs to be pollinated
The use of autonomous agriculture equipment, or robots, in greenhouse and orchards, and for fruit and vine crops, is well documented. What’s quickly catching up is robotic equipment in the use of broadacre crops, such as wheat, canola, soybean and other large acre crops in North America, as well as wheat, rape and vegetable crops in Europe, that are grown on larger tracts of land. These machines often focus specifically on tillage, seeding, spraying and harvesting, time-consuming tasks for large-area farmers.
Farmers continue to clamour for these robots at an increasing rate and it’s often for the simple reason that good help is hard to find. Labour shortages were already a challenge in countless nations before COVID-19 gripped the world. Today, that same supply crunch is only heightened. So, just how ubiquitous are these machines in broadacre, and what can they truly do? New AG International looked at a few companies leading the charge when it comes to autonomy.
Naïo Technologies — Toulouse, France In Toulouse, France, Naïo Technologies is slowly but surely moving from smaller scale into progressively larger autonomous equipment. Its first robot, Oz, is branded as a versatile farming assistant primarily designed to hoe, weed, seed and create furrows for small-scale operations. Its two other primary robots, Ted and Dino, work in larger scale vegetable operations and vineyards. Recently, though, the company is preparing to launch two new robots, Go and Orio, the latter of which is reserved for row crops including potatoes, sugarbeet and cereals. According to co-founder and CTO Gaëtan Severac, the new equipment is an extension not only of the company, but the types of work farmers have to deal with in such operations.
“Mechanical weeding control, it’s one of the most expensive tasks, and one of the more challenging tasks from the point of view of the farmer,” he says.
Turning weeding into an automated task makes sense to Severac as certain crop protection products may cause harm to humans if they are exposed, and in Europe’s tightly regulated marketplace, robots help solve this issue.
“The farmers find the best investment is really the mechanical control,” he says.
Severac also points out that if operators are stuck sitting in a tractor, it is logical to increase a machine’s size for maximum results, but robots again change all of that.
“If you are thinking with a human need for all tasks, you need a human to drive a tractor, it makes sense to have bigger and bigger machine. It’s the only way to augment the work of the human,” he explains. “If you think, ‘I do not need a human anymore,’ instead of one big machine, I can have five medium sized machines. It unlocks a lot of very interesting solutions again to improve environmental impact to have a lighter machine and less soil compaction.”
The company plans to introduce laser-driven weeding solutions within the next two to three years, as well.
Although robots were once considered a high-priced demonstration toy to some, Severac says the general sentiment towards autonomous agriculture continues to trend in a positive direction.
“Five, 10 years ago people were comparing robots to tractors, and saying, ‘for this price I can have a tractor, what’s the point?’ Now we do not hear such a comment. People are really seeing the value. It’s an investment. Buying a robot is a huge investment, but it’s worth it and it can bring a lot of value to farmers.”
AgXeed — Oirlo, Netherlands One of the younger companies in autonomous agriculture, AgXeed has worked at a frantic pace to bring technology to farmers. The robots created from the company are specialized in unusual, unclassifiable shapes of land and this is where AgXeed focuses itself.
“Most parts of the world, ag is not rectangular,” says Philipp Kamps, the company’s marketing manager. “A robot needs to adopt to the habitat of the farmer and not the other way around.”
Kamps says agriculture must get away from farmers on machines and free them up not be jammed completing repetitive, low-skill tasks.
With a simple slogan, “we provide autonomy,” the company’s robots link up to other traditional companies’ three-point linkage systems.
The response from growers: “They were pretty much delighted,” admits Kamps, adding that AgXeed collects zero farmer data by default. A farmer would have to specifically “allow the company to collect” information, which appears to be another reason the technology is attractive. It’s private.
The company’s robotic tractor is similar in cost and power to most brands’ 5-series offerings, but if you track the machine instead of wheels, that’s another cost, and they’re still not ready for autonomy, yet another cost. And the labour hasn’t even been calculated yet.
Because the robot is smaller than large North American market machinery, there is a smaller consumption of fuel and less soil compaction. AgXeed plans to reduce
the total cost of ownership by using smaller units with smaller implements. To achieve the same capacity, the robot just does more working hours, which wouldn’t be possible with human drivers, because they are just not available.
This spring, AgXeed will use a precision drill for corn, and in autumn, winter barley. The following year will see them move into hoeing, mechanical weeding and plowing. Based on work over the last four years since the company’s inception, Kamps is excited for the future and believes that autonomy and agriculture will be a perfect pair for years to come.
“Agriculture and environment, social and biological environments are growing together step by step. “Organic, conventional—everything is just growing together, and everyone is seeing it grow together.”
Raven Industries — South Dakota, U.S.
Although Europe may be ahead of the curve with its offerings, North America is not far behind and autonomous equipment does exist, albeit it typically more in beta stages. At Raven, a company recently acquired by CNHi, the organization is focused on providing autonomous ag solutions for farmers looking to retrofit existing machinery with autonomous capabilities through its OMNiDRIVE lineup or factory equipped machines through its OMNiPOWER lineup. Raven has deployed solutions in harvesting, spreading, spraying and seeding applications.
For North America in particular, labour challenges are plentiful in both Canada and the U.S. Farms that exceed 10,000 acres in size barely raise an eyebrow, but the workers are a precious commodity. Many farmers of scale are excited for self-driving machinery to help with spring seeding and harvest activities which are measured in weeks, not days, even with fully staffed crews.
“Consistent feedback is that the labour challenges growers are facing isn’t going away,” says Dominic Walkes, Raven’s director of strategic initiatives. “If anything, the COVID environment has exacerbated the situation.”
Farmers currently testing Raven products are able to control everything via a tablet through its mission control. That tablet is often attached to the inside of a tractor that a farmer is actually inside of, and it’s controlling another piece of machinery, perhaps a grain cart during harvest.
“The person combining can have complete control, hail it, and send it back to the edge of a field,” Walkes explains.
One area that Raven, and most North American companies face, is the challenge of rural connectivity. Unlike a highly condensed Europe, farms in Canada and the U.S. are extremely large and often have deep coulees and vast dead zones that immediately inhibit connectivity, vital for data collection as well as autonomous machinery communicating through IoT.
To combat this, Raven says its autonomous machines are built for such inevitable problems.
“It’s made to run in an offline mode essentially,” says Walkes. “Where we might not have reliable connectivity everywhere we go, we have some radio frequency solutions today. If it drops out for a minute or two, that’s not an issue, we can continue the operation.”
Raven’s machines collect data and Walkes says at this point they’re focused on farmers who are willing to share data and help the company create better products.
Today, no more than 100 American farmers are using Raven’s driverless products, including spreaders and sprayers alongside harvest solutions, but Walkes says that will start to scale in the coming years.
The North American marketplace as a whole is rapidly hitting a convergence
of customer readiness and product maturity, which is the perfect environment for a radical shift in how farmers do business. "We're really excited about what's to come," he says.
Labour challenges growers are facing isn’t going away
Olds College — Alberta, Canada Agricultural institutions continually try to teach and train the next generation about farming and agriculture. Whether it be agronomic best practices or understanding soil science and nutrients, there is a bedrock of information for all young men and women to learn. However, in 2022, the landscape has shifted where students are often digital-first problem solvers, and terms such as machine learning and autonomy are commonplace.
One such institution on the cutting-edge of agriculture is Olds College, located in Olds in the Canadian province of Alberta, located about 100 kilometres north of Calgary.
The college received a CDN$16 million philanthropic donation by provincial oilfield leader David Werklund in recent years to create a Smart Farm and new school within the college entirely dedicated to smart technologies. It also spurred the creation of new disciplines from the college, including a techgronomy diploma.
Joy Agnew, the college’s VP of applied research, says students have shown a tremendous capacity for learning all-new technologies within a short period of time through internship placements with the college’s formal research team. They work on and evaluate autonomous equipment and IoT. All students are trained on autonomous equipment and IoT as part of their regular course work.
“Summer students who join our team come in with zero experience and a little bit of traditional experience, and become skilled autonomous equipment operators at the end of their four-month internship,” says Agnew. “That’s what makes it so exciting for us.” From what Agnew has observed of autonomous equipment, it all comes down to labour measures, as well as in-field efficiencies.
“It’s a labour-saving device where you can have the equipment operating in the field and you’re observing but able to do other tasks or operate another piece [of equipment] at the same time,” notes Agnew. “We are already seeing and quantifying other environmental benefits, improved field efficiencies and autonomous equipment needs.”
Machinery being tested at the college is designed for use in the Americas and has the same wheelbase as manned machinery. But with less weight, it continues to reduce soil compaction and lower fuel consumption, a critical point as many countries – including Canada – are targeting net zero emissions standards by 2050, if not sooner. Results on broadacre crops are promising.
“We’ve experienced amazing progress in overall functionality and reliability of autonomous equipment operation on the Smart Farm, specifically the OMNi platform,” says Agnew. “In 2021, we seeded, sprayed and spread fertilizer on over 4,500 acres with the longest hands-off operating time of over five hours, compared to just over one hour in 2020. We are crunching the numbers now, but we expect to see improvements in overall field efficiency as well; and for some field configurations, we expect this efficiency to be better than with conventional equipment.”
Agnew does admit, though, that connectivity issues remain. Even where the college is located, only a few kilometres off the primary Highway 2 corridor connecting metropolitan centres of Calgary and the province’s capital city of Edmonton, there are still dead zones, even on the institute’s highly connected farmland.
Agnew says that even though Europe is usually one to two decades ahead of North America on ag tech, North America is rapidly catching up with robotics and autonomy.
“I can see it being adopted and becoming the norm,” she says. “The Canadian government is investing heavily in building and supporting the innovation ecosystem in Canada with a directed focus on ag and agri-food. This is helping to attract tech developers (and more investors) including autonomous ag tech to Canada, which is allowing Canada to build more expertise and training programs related to autonomous ag tech.” ●
Buying a robot is a huge investment, but it’s worth it and it can bring a lot of value to farmers
We expect to see improvements in overall field efficiency
Naïo Technologies’ first robot, autonomous robot, Oz, is branded as a versatile farming assistant. Photo: Naïo Technologies
AgXeed has worked at a frantic pace to bring technology to farmers. Photo: AgXeed
Raven is focused on providing autonomous ag solutions for farmers looking to retrofit existing machinery with autonomous capabilities.
Photo: Raven Industries
Olds College students are trained on autonomous equipment and IoT as part of their regular course work. Photo: Olds College
Erik de Vries, joint CEO, Agri Technovation speaks with Luke Hutson, Editor-in-Chief, New AG International
One of the winners of the Innovation Award at the Biostimulants World Congress in Miami in December 2021, Picklogger is a tool that goes beyond just linking a fruit to a particular orchard – it links a fruit to a particular tree. The other benefits extend far down the supply chain, explains Erik de Vries, joint CEO of Agri Technovation, the company that developed the device. Luke Hutson writes.
New AG International (NAI): Picklogger is a harvesting tool that records the location of every pick when a fruit is harvested. From what I saw at the Biostimulants World Congress in Miami in December 2021, the tool is a hand-held snipper that has a sensor that relays a signal each time it snips i.e., picks a fruit. The signal is then recorded – hence the name of the device – and the location of the pick is linked to the weight from the box or crate where the fruit was picked. In this way, Picklogger is recording not just the orchard, but also the tree where a fruit was picked. So, with this device you are extending the depth in the traceability within the supply chain. Is that essentially the concept behind Picklogger and how it works? Erik de Vries (EdV): Indeed. The GPS coordinates gathered with every snip are sent to the cloud and combined with weight data from the packhouse. These data points are the primary inputs required to deliver a fruit crop yield map, a first of its kind that we know of. This is a tool that generates big data with more than 220 million data points logged in the past season.
NAI: What other data is the tool gathering and how can it be used? If it is collecting the yield distribution from within an orchard, I’m thinking it can detect areas of lower yield and possibly alert the grower to areas of nutrient or water deficiency. EdV: Hitting the nail on the head here. Identifying in-orchard variance is one of the primary objectives for the Picklogger solution. The greatest benefit or value-add is that it allows the agriculturist, in cooperation with the producer, to home in on the variance within each of the orchards and accordingly develop corrective management strategies. The eventual result is increased harvest yields over time.
NAI: We also spoke how the linking to specific box or crate loads of fruit means that in the warehouse if there are any post-harvest issues with the fruit quality, the data from Picklogger can be used to go back to the orchard. Is that correct? EdV: Yes, by combining culling factor information from the modern packhouse systems back to the orchard via RFID tags from the box or crate to Picklogger in the orchard, valuable information regarding a range of quality parameters can be traced back to the orchard post-harvest.
NAI: What type of fruit trees is Picklogger used on, and could it be adapted for other crops? EdV: At the moment, it is primarily used on citrus, avocados, table grapes, mangoes and passion fruit. However, Picklogger ’s harvesting capabilities include other fruit that is harvested with shears, scissors or a pruner. Agri Technovation is also doing trials with vegetables like peppers (capsicums) and cucumbers.
NAI: On a technical note, how many of these snipping tools can the equipment handle in the field at any one time? EvD: Picklogger is fully scalable. The Picklogger device records data in the field during the harvest process and stores the data in local memory. When the device is then charged, normally during the night, it downloads the pick data to the cloud. The cloud servers also scale automatically based on demand, and the processed information is then represented on MyFarmWeb, an agricultural platform developed in co-operation with Vodafone, automatically.
NAI: Which part of the tool does Agri Technovation manufacture? And where have you sold the tool so far? Do you plan to export? EdV: Picklogger is fully assembled by Agri Technovation even though parts and components are sourced from more than 30 suppliers globally. Picklogger is actively running in the U.S., Peru, Spain, South Africa, Australia and New Zealand, with an ambition to onboard many more countries in the near future.
NAI: At the moment this is a hand-held tool for use with human pickers, but could it be adapted for robotics do you think? EdV: It certainly could. As a principle, Agri Technovation believes that a yield map is one of the most important data points on a farm. Combining the yield maps produced with other available agricultural information, such as soil classification, soil chemical analysis, leaf data, irrigation data, pest data, etc., enables well-informed decision-making and the formulation of corrective strategies. Picklogger is the method to get to that information for fruit crops harvested with a scissor, but any other device that produces a yield map for any other crop type, is something of interest to us.
NAI: What lies ahead in terms of future developments for Picklogger EdV: As Picklogger is launched in more countries across the world, the need to adapt for local requirements evolves. Currently the engineering team is working on the next version of the Picklogger that will integrate seamlessly with a cellular phone and use modern cellular phone charging options to recharge the battery. As mentioned previously, a lot of work is also being done to integrate with packhouses to extend range of data parameters generated by Picklogger. ●
You have the word ‘revolution’ in the title – why did you use that word?Revolutions are unexpected and their results and consequences are unpredictable. From my perspective precisely this is exactly how digitalization is currently affecting the agricultural industry nowadays. Hundreds of agrifood tech startups from all over the world are developing innovative new technologies and new business models challenging the way food is produced responding to a strong customer’s demand for a more efficient and more sustainable agriculture.
Digital technology in agriculture is such a huge topic – what aspects does your book focus on? One of the purposes of the book is to demonstrate how digitalization is affecting each one of the key aspects of agricultural business. Chapter by chapter, from breeding to commercialization, I analyze and present real cases showing and explaining how digitalization is revolutionizing the way food is produced. In the book hundreds of real life startups are mentioned and analyzed, showcasing how innovation is now coming from small and independent entrepreneurs.
What are the main predictions you explore in the book? The book’s main predictions is that agriculture is going to change in a dramatic way in the coming years. This change is not going to be simple, and it is not going to be easy for all key players, starting with the farmers but also impacting on key industry players. History demonstrates that when these drastic changes happen there is a huge impact on the business structure. As always happens, not everyone is ready to adapt to this new environment and new leadership should rise; meanwhile, old leaders should be challenged and eventually substituted. ●
Erik de Vries, Joint CEO, Agri Technovation
Silal, an Abu Dhabi fresh food and agritech company, and Hoogendoorn, a horticulture company, have launched the Digital Agronomy Service, with the aim of deploying IoT sensors in 100 farms over the course of 2022.
This project is aimed at enabling local farmers and advisors to make better decisions on irrigation, fertilization, and crop management to maximize locally grown fresh produce.
Through this collaboration, Silal will apply IoT sensors to capture key parameters affecting crop growth in greenhouses and net-houses, in order to optimize resource use efficiency and productivity. According to a news release, the initiative will help the company in determining the status of crop growth and input requirements of farms across the emirate, to devise prudent agriculture plans and projects. For farmers, this initiative will help to drive farm productivity, produce quality and profitability.
"The agricultural sector has undergone several revolutions, perhaps the most transformative of which is the use of agritech,” said Salmeen Obaid Alameri, CEO of Silal. “Wwe believe that digital transformation is key to enhance agricultural sustainability practices while empowering local farmers and helping them increase their crop yields."
According to Martin Helmich, CCO of Hoogendoorn Growth Management, local production is becoming increasingly important, globally and especially in the Middle East. “We witness enormous interest in technology and knowledge to increase local fresh produce and efficient use of resources and inputs in this region,” he noted. “Therefore, Hoogendoorn has committed to working with Silal to help the local growers, and we will deploy our IoT technologies, advance data analytics and crop knowledge to create world-class agronomic decision support service to support local food production in the UAE."
This initiative will deploy a wide range of IoT sensors to capture key parameters, such as temperature, humidity, vapour difference, radiation, pH, soil moisture and electric conductivity. The sensors will also determine the status of crop growth and input requirements, such as water, energy, CO2, fertilizers and agrichemical applications. These data will be connected to AI-powered computers and used by Silal’s agriculture engineers to devise crop growth models, thereby making better agronomic decisions and increasing productivity. ●
Carbon Robotics, an agricultural robotics company, unveiled its 2022 LaserWeeder implement, an autonomous, laserweeding pull-behind robot that attaches to the back of tractors.
The updated LaserWeeder features 30 industrial CO2 lasers, more than three times the lasers in Carbon Robotics’ self-driving autonomous LaserWeeder, creating an average weeding capacity of two acres per hour.
Carbon Robotics worked closely with leading vegetable growers to design the 2022 implement so it integrates effortlessly into existing farming infrastructure while covering more ground and solving problems associated with spraying, hand weeding and mechanical weeding.
The company stated that this new LaserWeeder has an average effective weeding capacity of two acres per hour. It is fully adjustable for crop row widths ranging from 60 to 84 inches, and adjustments for transitioning between different crops can be made via a touchscreen. The LaserWeeder is towed by common row tractors with a three-point hitch. And, this 2022 version, like its self-driving 2021 predecessor, features Carbon Robotics’ artificial intelligence technology that enables the robot to instantly identify, target and eliminate weeds using thermal energy. It can be operated day or night in all weather conditions.
To scale manufacturing of its implement, the company raised a $27 million Series B in 2021. Carbon Robotics was founded in 2018 and is based in Seattle. Carbon Robotics is accepting pre-orders for 2023. ●
Olds College (Alberta, Canada) is partnering with Wyvern to improve sustainability in agriculture using high-resolution hyperspectral imagery with the upcoming launch of Wyvern’s DragonEye satellite.
In January, Wyvern said it had raised USD$4.5 million through its pre-seed and seed funding rounds. Then in
February, Wyvern — a Canadian satellite startup company in Edmonton (Alberta, Canada) — received a CDN$4 million investment from Sustainable Development Technology Canada (SDTC) which will help Wyvern launch their DragonEye satellite. This satellite will deliver high-resolution hyperspectral imagery from high-quality camera technology to help improve on-farm management.
Wyvern’s proprietary deployable optics technology is key to their ability to deliver affordable high resolution hyperspectral imagery, including 1m VNIR and 5m SWIR, in the coming years. DragonEye will be Wyvern’s first satellite equipped with this innovative technology that unfolds a telescope in space, similar to the James Webb telescope launched in late 2021.
Wyvern, Olds College, and numerous other collaborators, including xarvio Digital Farming Solutions, SkyWatch, Metaspectral, and Wild + Pine, are working together on the three-year project to collect and use data from Wyvern’s DragonEye satellite to improve the overall efficiency of crop inputs — which includes helping farmers use less fertilizer, pesticides and water along with producing bigger yields. ●
Wyvern is designing telescopes that are compact on launch and deploy in space.
Image: Wyvern
A world-first service that could revolutionize the way satellite imagery is used in precision agriculture was announced last month by UK technology companies Origin Digital and Aspia Space.
The new ‘ClearSky’ service, launching imminently in the UK, feeds radar data into a deep neural network to derive the view of a field that a satellite would see if there were no clouds blocking its camera.
This innovation means that farmers using ClearSky are guaranteed to receive an image every six days showing them how their crop is developing, whatever the weather. This is in contrast to traditional, weather-dependent imagery which can often have gaps of several weeks between cloud-free views.
“ClearSky is a hugely exciting development that we’re delighted to bring to UK agriculture, because it guarantees the ingredient of dependable regularity that precision farming systems need to deliver optimal results, but which has been missing from traditional imagery services,” said Madhumita Mund Rao, head of data at Origin Digital.” This reliability will give UK farmers a substantial new advantage in sustainably optimizing their yield and input use.”
At any given time, an average of 67 percent of the Earth is covered by clouds, so precision agriculture systems that rely on getting clear satellite imagery at the right time have historically struggled to deliver on their high potential value. “ClearSky eliminates that struggle by guaranteeing the consistent regularity these systems need to deliver results, enabling farmers to fully optimize their fertilizer use, for example, and helping both their wallets and the planet,” Rao added.
Analysis by Origin Digital shows that the widely used European Space Agency ‘Sentinel 2’ satellites produced 13 clear images per UK farm on average in 2021. In contrast, the ClearSky technology developed by Aspia Space uses revolutionary techniques to produce more than 60 cloud-free images per year, which can be used alongside the clear images captured by Sentinel 2 and other providers.
According to Aspia Space co-founder, professor Jim Geach, Aspia’s technology unlocks Earth observation imaging data and intelligence that would have otherwise been lost. “ClearSky uses radar inputs, which penetrate cloud but are challenging to interpret, to derive imagery across the visible and shortwave infrared spectrum. This means that even in the presence of 100 percent cloud cover, we can deliver regular, reliable and consistent cloud-free images that are easily understood and can be analyzed in exactly the same way as regular optical imagery,” he said.
“ClearSky was developed using the idea that the way radio and microwaves behave when they hit surface features – such as crops – is correlated, albeit in a highly complex way, with the way that optical light waves interact with those same features. Using AI to unpick this correlation means that ClearSky can predict cloud-free imagery with no optical inputs without a loss of accuracy over long periods of time without clear optical images.”
Origin Digital and Aspia Space plan to deepen their collaboration to localize and export the benefits of ClearSky to farmers around the world, as well as developing further potential applications that bring innovative data insights to UK agriculture. ●
Images of the Humber region, UK, from Sentinel 2 (left) and Aspia Space (right). Image: Aspia Space
