The face of agriculture, as we know it, is changing. Technology is playing a pivotal role in making farming techniques smarter than ever before (which is mighty important, given the spiralling food requirements of the burgeoning global population). Switching over to smart precision farming techniques is enabling crop-growers to significantly increase productivity levels from their lands (i.e., making more with the same resources). The role of robotics, powered by powerful GPS technologies and big data, in agriculture is firmly in focus at present. By 2022, the worldwide agri-robots (or, Agbots) market will go beyond $12.50 million – ~365% more than the corresponding figure in 2016. The CAGR for the 2017-2022 period will be just a touch under 21%.
Thanks to the rapid evolutions in farmtech over the last half a decade or so, agricultural robots are no longer about only the John Deere smart tractors and agricultural drones/UAVs. North America and Europe lead the way in precision farming and usage of farming robots, with Asia-Pacific (primarily, China and India) also growing at a fast clip. The growing awareness about IoT (internet of things) in general, and smart farming techniques in particular, is helping end users optimally use agricultural robots – for a myriad of purposes, speeding up processes, making things efficient, and ruling out chances of manual errors. According to a recent Tractica report, annual shipments of such robots will touch the 594000 units mark by the end of 2024 (the annual shipment in 2016 was 33000 units). In what follows, we will focus on some important functions and use cases of robots in farming:
Types of farming robots
As the implementation of data-backed, tech-based agriculture is rising, more and more types of agbots are coming into the picture. Robots with advanced artificial intelligence (AI) capabilities and built-in analytics systems are being used for various on-field tasks – right from crop and cattle management, to dairy management, soil monitoring, and overall farm yield optimization. The watchword here is sustainable expansion of agricultural produce, with the help of technology. Apart from agricultural drones and GPS-powered smart tractors, milking robots, unmanned spraying helicopters and materials management systems are steadily growing in popularity. Water management and irrigation is yet another domain where agbots are delivering considerable benefits. There are ‘smart harvesting robots’ as well, as well as ‘intelligent’ tools for farm inventory management. In a nutshell, robots are giving shape to the concept of fully ‘digital farms’. Interestingly, agbots have also provided a much-needed thrust to indoor farming practices.
Note: Clearpath Robotics, John Deere, AGCO and Lely are some of the biggest players in the worldwide farming robots industry.
Micro-level crop monitoring
In the United States, the average size of agricultural farms is 444 acres (this figure has remained relatively stable over the last couple of decades). Now, it is extremely difficult, if not impossible, to manually monitor large fields. Lack of actionable data and insights, in turn, increase the uncertainty factor associated with farming – and to mitigate such risks, agbots (ground-level robots) and drones can play a very important role. These farm robots typically make use of powerful sensors and geomapping technologies to bring holistic, real-time crop information to end-users (i.e., the crop growers). There are certain agbots, like BoniRob, which can reach very close to the ground crops – in order to deliver granular-level crop information. Several companies have also started offering farming hardware and software analytics tools as a package.
Note: Broadly speaking, agricultural robotics system have 5 key components – the cloud network, the satellite system, the actual farm robot, the smartphone/tablet, and the logistic unit.
Micro-spraying and weeding
In a 2016 report, it was estimated that QUT’s Agbot II had the capacity of pulling up the annual savings of Australian farmers to $1.3 billion. Micro-spraying and weeding robots come as a welcome alternative to spreading pesticides on fields manually – which: a) is unduly time-consuming, b) leads to wastage/overuse of chemicals, and c) might have adverse effects on the environment. Agbots used for weeding are powered with advanced computer vision – for identifying the weed-affected areas correctly, and spraying the required amounts of pesticide on those areas. This, understandably, reduces the overall use of herbicides – helping farmers cut down on unnecessary expenses. Robots that spot and uproot weeds (instead of applying chemicals on them) can also be used. On average, autonomous weeding techniques with agbots can kill ~90% of pests, saving 75% of pesticide usage simultaneously.
Note: Laser technology can also be used by certain agri robots to kill weeds. Accurate tactile sensing capabilities is a must-have in any good weeding/micro-spraying robot.
The role of drones and sensors
By January 2017, the value of the global agri-drone market had swelled to $32.5 million (as per a PwC report). Land mapping, on-field pest detection and general crop-inspection are some important purposes for which drones are being extensively used. In fact, we are probably not far away from a time when agri-drones would be able to communicate with each other and ‘work as a team’ (say, for in-depth crop treatment, irrigation management, and weeding). In the smart agriculture ecosystem, sensors also play a crucial role . For instance, soil moisture sensors can constantly track the condition of the soil (moisture levels, humidity, etc.), send the data to the cloud network, and generate irrigation notifications for the farmers. As a result, chances of water wastage are done away with. In New Zealand, SmartN is already being used to track the places where cows have urinated – since such areas do not need any further fertilizer or chemicals. Without properly functioning on-field drones and sensors, the utility of agricultural robots would have been very limited.
Note: The agricultural drone market will be worth $2.9 billion by the end of 2020 – nearly 5X more than the corresponding figure in 2015.
More efficient picking, sorting and harvesting
The European Union-funded cROPS project (Clever Robots For Crops) have shown the way for this. Dedicated applications for picking and sorting ‘soft fruits’ are being tested and deployed – and these applications typically have the degree of manual dexterity required for the job (crops like wheat and corn can be easily harvested by automated combine harvesters). Octinion, a leading research company, released a prototype of its strawberry-picking robot a few months back. The robot can sort and pick 70% of all ripe strawberries (Shibiya Seiki – a Japanese company – showcased a similar agbot in 2013). Autonomous apple-picking robots, armed with vacuum removal mechanics and computer vision (so that the tree itself, or the fruit, is not damaged in any way) are coming into the picture – and they are expected to become fairly mainstream in the next couple of years or so. The vision systems in the agbots enable the latter to gauge the ripeness of the fruits, detect the presence of dust on them, and manage temperature/wind conditions. Crop damage at the time of harvest is a big issue in traditional agriculture – and farming robotics is designed to eliminate that concern.
Note: Abundant Robotics (California, USA) and FF Robotics (Israel) are both conducting trial runs of their apple picker robots. The built-in AI sensors can also let farmers know of the correct harvesting time/window for their crops.
Key drivers of farming with robots
Come 2024, we will be looking at annual revenues in excess of $74 billion from the global agricultural robots market. The exponentially rising global population is, arguably, the single biggest factor behind the rapid developments in precision farming with robots – since there are more mouths to feed, total land resources are limited, and yields have to be maximized. Using agbots and smart farming techniques are also likely to generate significant economies of scale in larger farms. The fact that wage bills make up nearly 41% of the total farming expenses is also motivating farm-owners to replace a section of the labor with agbots. Unavailability of sufficient manpower is yet another factor – with the youth mostly uninterested in getting involved in traditional farming and getting their hands dirty (literally) in the process. The growing food scarcity levels, the constant climatic changes and farmland transfers, and the existing bottlenecks of manual farming (along with the uncertainties and low yields) are also making people switch over to agbots as viable (and much more efficient) alternatives. The impact of IoT in agriculture is growing all the time, and robotic farming offers manifold advantages.
Note: Not all is plain sailing for agricultural robots yet though. The technology itself is rather fragmented, there is still a widespread lack of awareness among farmers, there are infrastructural issues, and the benefits/value propositions of robots in farming might not be immediately apparent. As farmtech standards evolve further, these issues will gradually get resolved.
Planting and seeding with agbots
In India, nearly 16% of all fruits and vegetable produce was lost in 2016-2017. The waste percentage of cereals was also alarmingly high. Across the world, a lot of wastage occurs at the time of seeding – simply because farmers follow the outdated method of spraying seeds from a moving tractor (with a so-called ‘broadcast spreader’). Farm robots with cutting-edge geomapping functionality can be used as an alternative – to prevent seeds from being wasted. These agbots are designed in the form of ‘robotic seeding attachments’ (to be attached to tractors) – and they can accurately determine all relevant soil features, so that the right seeds are dispersed at the right places in the right time. Agbots can also power nursery automation – ensuring that all greenhouse activities, like potting, seeding and warehousing, are optimized. Precision seeding, in a way, is all about maximizing the chances of growth of food plants from seeds.
Note: With urbanization happening globally at a rapid clip, agbots are helping in the creation, management and maintenance of indoor farms (in urban environments). In Hong Kong, such a farm robotic system can bring down water usage by as much as 95%, while the growth of plants also gets considerably accelerated.
Rise and rise of fully unmanned technology with robots
Smart tractors have been in existence for a long time – but certain factors are still holding the technology back. Primary among them are the relatively high price tags of sensors (for small farmers, this is a huge factor), a feeling of distrust stemming from lack of knowledge, and regulatory issues. Going forward though, agricultural robots certainly has the potential of ushering in fully unmanned farm technologies. The total number of smart tractor shipments (with GPS guidance, or some other form of steering technology) is expected to be >700K in 2028 – and by 2038, more than 40000 Level 5 tractors will be sold. Agricultural robots will gradually start to work in fleets (rather than as standalone units) – and most of them will have at least some degree of functional autonomy. According to experts, 2024 will be ‘trigger year’ for agbots – following which shipments will start to accelerate in a big way. The cost of farm automation suites will also start to fall gradually, further fueling adoption levels. ‘Intelligent agribots’ are not going to totally replace humans on agricultural fields – but they will be the face of autonomous precision farming in the foreseeable future.
Note: Agribotix, a noted agricultural intelligence company, reported in a case study that it had managed to reduce crop losses by 13%, on a 110-acre land.
Robotics for milking
By 2017 Q3, more than 34000 robotic milking systems (or, RMS) were in active use on farms across the globe. Instead of having to deal with risk of infections and low volume of milk production – collaborative agbots can be used to sprinkle ‘safe’ disinfectants on the animals’ udders, prior to the milking process. What’s more – these robotic solutions also ‘prepare’ the animals for milking – which ensures that the volume and quality of milk obtained is optimal. As a rule of thumb, the frequency of milking has to steadily decrease over the lactation cycle. The high price of milking robots (~$175000 for a system that can milk 50-70 cows) is a concern – but over time, it can be reasonably expected that the cost will come down. With proper management and regulated operations (training the workforce will also be crucial), cow-milking robots can yield definite advantages – which may not be apparent at first.
Note: It has been found that, in the first three days of the robotic milking process, 3 out of every 4 cows voluntarily go to these robots.
10. Mowing, Pruning and Thinning
What should be the ideal spacing between seeds on a field – so that the chances of healthy crop growth would be optimal? Traditionally, experience and guesswork were the only tools to answer this question. The scenario has now changed though, with specialized agbots having the capability of gauging the ‘correct’ density of plants, for proper growth (the robots typically reduce the density). This is known as ‘thinning’ – and even with robots, the process can take quite a bit of time. Agri-robots are also being used to mow or cut certain portions of plants, to foster their healthy growth (this is known as ‘pruning’). For both ‘thinning’ and ‘pruning’ crops in the best possible manner, agbots make use of high-level computer vision technology. Automated pruning has already been done in the wine industry (vineyard robot Wall-Ye) – and a robot for blueberry pruning has also been developed.
Note: Last year, LettuceBot – created by Blue River Technologies – bagged an ‘outstanding product innovation in agriculture’ award.
11. Technical refinements are required
The benefits of switching over to smart agriculture and robotic farming are pretty much well-documented. Even so, issues remain – primarily from a technical perspective – in the path of the growth of agritech in general, and agbots in particular. For starters, the lack of regulatory uniformity is a big problem for both the OEMs as well as the final users. There are also scopes for improving the accuracy of robot positions (which, in turn, will make weeding, irrigation, crop monitoring, and other activities more effective). Question marks still remain over the safety factor of farm robots, for humans as well as for the environment. In addition, it is also uncertain as to who would shoulder the responsibilities in case an unforeseen accident does occur. For accurate analysis (for example, determining the ripeness of fruits before picking), robots need to factor in external things like temperature fluctuations and light variations. The investments required for deploying agbot systems are large – and unless the technology is properly refined – they won’t be able to deliver good value for money. There’s still a long way to go.
Note: Precision irrigation is rapidly growing popular across the globe. It offers two important advantages over traditional irrigation methods – firstly, areas that were previously inaccessible can now be reached, and secondly, plants can be targeted separately.
12. Sheep herding with robots
Shepherding cattle herds is one of the oldest, and also one of the trickiest, tasks associated with cattle farming. Agricultural robots, in the form of drones, can automate this process as well. Already, UAVs are being used in Ireland (copters) and New Zealand to manage sheep herds. These drones can, of course, be remotely operated – and their constant tracking operations make sure that not a single animal strays away from the herd. Another interesting point is that the sheep followed the drones out of their own free will. The flipside to this is the short battery life of these cattle drones (~20-25 minutes) and their steep price (the Q500 drones are available at $1299). In Australia, many ranchers use helicopters for cattle herding. Drones can be an excellent, and a much cheaper, alternative for them.
Note: The practice of fertilizer application on agricultural fields has also been revolutionized, thanks to the advent of farm robots. The robotics systems of Rowbot, for instance, ensure effective utilization of nitrogen fertilizer.
Present-day farming is – or at least should be – all about the ‘sustainability of agriculture’. Agricultural robots, for all their advantages and powerful functionalities, are still at a nascent stage – and there are still several rough edges. As these issues get resolved, robots in farming would start to increase annual yields, lower crop/food prices, ensure better food accessibility levels, and make farming standards more efficient than ever before.
Agriculture is increasingly becoming tech-based – and the impact of IoT on farming has been profound. With the need for improving agricultural productivity levels increasing everyday, it can safely be assumed that agbots will become more mainstream in the coming years. It also has to be kept in mind that these robots are not, and will never be, meant for working in isolation. A certain level of human interaction (for management, operation, performance tracking, etc.), will always be required.
Automated farming is definitely the future. Agri-robots will be right at the face of it in the next decade.
Latest posts by Hussain Fakhruddin (see all)
- AI-as-a-Service: All That You Need To Know - October 30, 2018
- Top 12 Trends In Digital Transformation To Watch Out For In 2019 - October 23, 2018
- Machine Learning in 2019: Tracing The Artificial Intelligence Growth Path - October 9, 2018