Drones, specifically quadcopters, are an adaptable lot. They’ve been used to assess damage after disasters, deliver ropes and life-jackets in areas too dangerous for ground-based rescuers, survey buildings on fire and deliver medical specimens.

But to achieve their full potential, they have to be tough. In the real world, drones are forced to navigate uncertain shapes in collapsing buildings, avoid obstacles and deal with challenging conditions, including storms and earthquakes.

At the USC Viterbi School of Engineering’s Department of Computer Science, researchers have created artificially intelligent drones that can quickly recover when pushed, kicked or when colliding with an object. The autonomous drone “learns” how to recover from a slew of challenging situations thrown at it during a simulation process.

“Currently, the controllers designed to stabilize quadcopters require careful tuning and even then, they are limited in terms of robustness to disruption and are model-specific,” said the study’s lead author Artem Molchanov, a Ph.D. in computer science candidate in USC’s Robotic Systems Embedded Laboratory.

“We’re trying to eliminate this problem and present an approach that leverages recent advancement in reinforcement learning so we can completely eliminate hand-tuning controllers and make drones super robust to disruptions.”

The paper, called “Sim-to-(Multi)-Real: Transfer of Low-Level Robust Control Policies to Multiple Quadrotors,” was presented at the International Conference on Intelligent Robots and Systems.

Co-authors were Tao Chen, USC computer science master’s student; Wolfgang Honig, a former USC computer science Ph.D. student; James A. Preiss, a computer science Ph.D. student; Nora Ayanian, USC assistant professor of computer science and Andrew and Erna Viterbi Early Career Chair; and Gaurav Sukhatme, professor of computer science and electrical and computer engineering and USC Viterbi executive vice dean.

Learning to fly

Roboticists have been turning to birds for flight inspiration for years. But drones have a long way to go before they’re as agile as their feathered counterparts. When a drone ends up in an undesirable orientation, such as upside down, it can be difficult for it to right itself. “A drone is an inherently unstable system,” said Molchanov.

“Controlling a drone requires a lot of precision. Especially when something sudden occurs, you need a fast and precise sequence of control inputs.” But, if a drone was able to learn from experience, like humans, it would be more capable of overcoming these challenges.

With this is mind, the USC researcher team created a system that uses a type of machine learning, a subset of artificial intelligence, called reinforcement learning to train the drone in a simulated environment. More precisely, to train the drone’s “brain,” or neural network controller.

“Reinforcement learning is inspired by biology—it’s very similar to how you might train a dog with a reward when it completes a command,” said Molchanov.

Of course, drones don’t get snacks. But in the process of reinforcement learning, they do receive an algorithmic reward: a mathematical reinforcement signal, which is positive reinforcement that it uses to infer which actions are most desirable.

Learning in simulation

The drone starts in simulation mode. At first, it knows nothing about the world or what it is trying to achieve, said Molchanov. It tries to jump a little bit or rotate on the ground.

Eventually, it learns to fly a little bit and receives the positive reinforcement signal. Gradually, through this process, it understands how to balance itself and ultimately fly. Then, things get more complicated.

While still in simulation, the researchers throw randomized conditions at the controller until it learns to deal with them successfully. They add noise to the input to simulate a realistic sensor. They change the size and strength of the motor and push the drone from different angles.

Over the course of 24 hours, the system processes 250 hours of real-world training. Like training wheels, learning in simulation mode allows the drone to learn on its own in a safe environment, before being released into the wild. Eventually, it finds solutions to every challenge put in its path.

“In simulation we can run hundreds of thousands of scenarios,” said Molchanov.

“We keep slightly changing the simulator, which allows the drone to learn to adapt to all possible imperfections of the environment.”

A real-world challenge

To prove their approach, the researchers moved the trained controller onto real drones developed in Ayanian’s Automatic Coordination of Teams Lab. In a netted indoor drone facility, they flew the drones and tried to throw them off by kicking and pushing them.

The drones were successful in correcting themselves from moderate hits (including pushes, light kicks and colliding with an object) 90% of the time. Once trained on one machine, the controller was able to quickly generalize to quadcopters with different dimensions, weights and sizes.

While the researchers focused on robustness in this study, they were surprised to find the system also performed competitively in terms of trajectory tracking—moving from point A to B to C. While not specifically trained for this purpose, it seems the rigorous simulation training also equipped the controller to follow a moving target precisely.

The researchers note that there’s still work to be done. In this experiment, they manually adjusted a few parameters on the drones, for example, limiting maximum thrust, but the next step is to make the drones completely independent. The experiment is a promising move towards building sturdy drones that can tune themselves and learn from experience.

Professor Sukhatme, Molchanov’s advisor and a Fletcher Jones Foundation Endowed Chair in Computer Science, said the research solves two important problems in robotics: robustness and generalization.

“From a safety perspective, robustness is super important. If you’re building a flight control system, it can’t be brittle and fall apart when something goes wrong,” said Sukhatme.

“The other important thing is generalization. Sometimes you might build a very safe system, but it will be very specialized. This research shows what a mature and accomplished Ph.D. student can achieve, and I’m very proud of Artem and the team he assembled.”

The post Advanced series of more robust drones are teaching themselves how to fly appeared first on Artificial Intelligence.

On Feb 3, medical workers at Wuhan Pulmonary Hospital in the central province of Hubei welcomed a new “comrade”, one who had no fear of the novel coronavirus even though he wasn’t wearing protective clothing － a robot designed to deliver medication and meals.

“Please wear masks and wash your hands frequently during this special period. Let’s conquer the difficulties together,” the robot intoned as it moved around the Department of Respiratory Medicine.

Li Ming, deputy director of the hospital’s information department, said the robot’s route is based on a map set by a computer programme.

Nurses bring meals for the 60 patients to the quarantine area and place them on the robot, which can deliver 10 to 15 in one journey, saving time and energy while also reducing human contact.

In a bid to save protective gear and prevent infections among medical workers, the hospital has launched an online system that allows doctors to check patients via video screens.

Meanwhile, an online diagnosis system has been built to support treatment in hospitals in Enshi and Xiangyang, both in Hubei, and the expert team at Wuhan Pulmonary Hospital has provided diagnoses for more than 30 patients via remote consultations.

UPGRADES

Faced with the epidemic, governments at all levels, tech companies and local communities are joining the battle against the coronavirus by upgrading the technological aspects of control and prevention work.

Last month, President Xi Jinping said the battle against the epidemic cannot be won without scientific and technological support, according to Xinhua News Agency.

On Feb 28, the Ministry of Science and Technology published a notice on its website encouraging businesses to deploy robots, temperature-screening machines and other devices that could help to reduce human contact as the nation returns to work following the enforced layoff after the Spring Festival Holiday.

Big data is being collected on platforms to monitor the development of the epidemic, while drones are being used to disinfect roads and open spaces. Meanwhile, temperature-screening technology and artificial intelligence are being employed in the detection and analysis of possible risks amid the gradual resumption of work.

Residents nationwide can trace the latest developments of the epidemic and obtain health tips from online sources, such as WeChat mini programs launched by internet companies like Tencent Holdings.

An epidemic map, locations of communities with confirmed cases and information to refute rumours are available on the monitoring platforms, which aim to provide people with accurate information in a timely manner.

Cities in Hubei, such as Xiangyang, Suizhou and Xiaogan, have built online platforms where people can report developments related to the epidemic, search their travel history to see if they shared planes or rail carriages with people who were later confirmed as having the coronavirus, and receive a remote diagnosis.

Led by the local governments of several cities in Hubei and the e-commerce giant Alibaba, these kinds of platforms facilitate control-and-prevention work in communities by ensuring that people returning from Wuhan are registered and placed under observation.

SOFTWARE SOLUTIONS

Alipay Health Code － smartphone software that indicates health status － was rolled out in more than 100 cities nationwide within a week of being introduced in Hangzhou, Zhejiang province, on Feb 11, according to Xinhua News Agency.

After people fill in a survey of their recent travel history, close contacts, temperature and related symptoms, they are assigned a code in green, yellow or red, depending on their suggested physical condition.

Those with green codes can freely enter or exit communities and workplaces, while those with red or yellow codes should enter a period of self-quarantine.

As of Feb 28, more than 15 million people in Zhejiang had registered with the system, which provides a visual indication of whether it is safe for them to return to work. “Have you turned green?” has become a popular catchphrase in the province.

NONHUMAN DELIVERIES

In addition to robots in hospitals, the e-commerce giant JD is deploying self-driving robots to deliver items to medical workers in Wuhan in a bid to protect both customers and employees.

The robots can easily cover the distance between the logistics point and a designated hospital, about 600 meters, according to the Shanghai Observer news portal.

Meanwhile, in some parts of the country, drones are being used to patrol open places, spray disinfectant and conduct thermal-imaging procedures.

Liu Shuai, a farmer in Mianzhu, Sichuan province, allowed the local village authorities to use a newly purchased agricultural drone to spray disinfectant.

“Drones are more convenient, and they can spray material more evenly than manual or electric devices,” he told Red Star News.

He volunteered to operate the drone and spray roads in his village at 9 am every morning. The work used to take a team a whole day to finish, but Liu completes the task in about three hours.

Meanwhile, a community official in Yichun, Jiangxi province, started a trend of using drones equipped with thermal-imaging technology to assess body temperatures, according to news portal Jiemian.

Yi Jinyu, who enjoys working with new technologies such as drones, 3D printing and data visualisation, bought the device from DJI, a leading provider of easy-to-fly drones.

During its 30 minutes of flight time per charge, the drone can detect temperatures ranging from -10 deg C to 140 deg C, thereby meeting all the demands for widespread inspections of people and goods in open spaces.

In January, Yi came up with the idea of deploying drones to assess temperatures.

After he started using the drone to gauge temperatures, many other communities quickly decided to follow suit.

On Feb 2, a community in Hangzhou started using the devices to conduct temperature checks in open spaces, such as parks.

If an individual registers a temperature reading that may indicate a fever, he or she is located by officials who conduct tests to ascertain the details of the situation and respond accordingly.

The post Robots, drones, AI and big data bring modern touch to disease control in China appeared first on Artificial Intelligence.

A version of this article was originally published on Medium. The views expressed here are solely those of the authors and do not represent positions of IEEE Spectrum or the IEEE.

We here at Skydio have been developing and deploying machine learning systems for years due to their ability to scale and improve with data. However, to date our learning systems have only been used for interpreting information about the world; in this post, we present our first machine learning system for actually acting in the world.

Using a novel learning algorithm, the Skydio autonomy engine, and only 3 hours of “off-policy” logged data, we trained a deep neural network pilot that is capable of filming and tracking a subject while avoiding obstacles.

We approached the problem of training a deep neural network pilot through the lens of imitation learning, in which the goal is to train a model that imitates an expert. Imitation learning was an appealing approach for us because we have a huge trove of flight data with an excellent drone pilot—the motion planner inside the Skydio autonomy engine. However, we quickly found that standard imitation learning performed poorly when applied to our challenging, real-world problem domain.

Standard imitation learning worked fine in easy scenarios, but did not generalize well to difficult ones. We propose that the signal of the expert’s trajectory is not rich enough to learn efficiently. Especially within our domain of flying through the air, the exact choice of flight path is a weak signal because there can be many obstacle-free paths that lead to cinematic video. The average scenario overwhelms the training signal.

How can we do better? Our insight is that we don’t have just any expert, we have a computational expert: the Skydio Autonomy Engine. Therefore instead of imitating what the expert does, we understand what the expert cares about. We call this approach Computational Expert Imitation Learning, or CEILing.

Why is CEILing better than standard imitation learning? Let’s consider a didactic example in which a teacher is trying to teach a student how to do multiplication. The teacher is deciding between two possible lesson plans. The first lesson plan is to give the student a bunch of multiplication problems, along with the answer key, and leave the student alone to let them figure out how multiplication works. The second lesson plan is to let the student attempt to solve some multiplication problems, give the student feedback on the exact mistakes they made, and continue until the student has mastered the topic.

Which lesson plan should the teacher choose? The second lesson plan is likely to be more effective because the student not only learns the correct answer, but also learns why the answer is correct. This allows the student to be able to solve multiplication problems they have never encountered before.

This same insight applies to robot navigation: Some deviations from the expert should be penalized more heavily than others. For example, deviating from the expert is generally okay in open space, but a critical mistake if it is towards an obstacle or causes visual loss of the subject. CEILing lets us convey that information from the expert instead of blindly penalizing deviations from the expert’s trajectory. This is why CEILing trains a deep neural pilot that generalizes well with little data.

Now one question to ask is why even use CEILing to train a deep neural pilot? Why not just have the computational expert be the pilot? The primary reason we are excited about CEILing is that CEILing could train a pilot that is actually better than the computational expert pilot.

How is this possible? Consider a scenario in which a drone needs to fly through a forest at high speed. This is a challenging environment because thin objects, such as tree branches, are difficult to see from far away. Although the current Skydio autonomy engine is able to perceive and avoid these thin branches, sometimes the branches can only be detected when the drone is already quite close, which forces the drone to execute an aggressive maneuver. In contrast, our deep neural pilot could be able to smoothly avoid these thin branches altogether because it will have learned that flying towards trees—which are large and easily seen—is correlated with encountering thin branches. In short, CEILing can leverage acausal (future) data, which enables it to “see” farther into the future and therefore train an even smarter pilot.

Although there is still much work to be done before the learned system will outperform our production system, we believe in pursuing leapfrog technologies. Deep reinforcement learning techniques promise to let us improve our entire system in a data-driven way, which will lead to an even smarter autonomous flying camera.

The post How We Trained a Deep Neural Pilot to Autonomously Fly the Skydio Drone appeared first on Artificial Intelligence.

The UK’s National Health Service continues to suffer the longest funding squeeze since it was established 71 years ago. That financial pressure has resulted in the service missing targets for how soon cancer patients should be referred for treatment for the past three years and waiting times in Accident and Emergency departments being at record levels.

Such is the financial and staffing pressure on the service, that talking about how recent advances in artificial intelligence (AI) could be applied to the NHS might seem fanciful.

Yet Professor Tony Young, national clinical director for innovation at NHS England, believes healthcare is at an inflexion point, where machine-learning technology could fuel huge advances in what’s possible.

“I think that healthcare is heading for one of those giant-leap moments in the next five to 10 years and AI is going to be a key tool in enabling us to take that giant leap,” he said, speaking at an event in London organized by The King’s Fund and IBM Watson Health.

Young highlighted technologies ranging from drones to robotics being considered for or used in the NHS, all of which take advantage of recent breakthroughs in machine learning – a field of AI.

ANALYZING BRAIN SCANS OF STROKE VICTIMS – VIZ.AI

One of the first companies to win FDA approval for an AI algorithm, according to Young, Viz.ai offers a system that scours images from CT (Computerized Tomography) scans of the brain for signs of damage following strokes.

“When you’ve got a regional stroke centre you might say, ‘Which CT do I need to look at first?’, and you’ll work your way through the pile, but there’s someone in there who you could save their brain if you intervened now.”

The system helps triage the scans, highlighting to doctors those patients who need urgent treatment, says Young.

Like much machine learning-powered technology in healthcare, Young says the company behind the system was co-founded by a doctor, a pediatric orthopedic surgeon at St Bartholomew’s Hospital in London.

SPOTTING BRAIN TUMOURS – MICROSOFT

Microsoft is working with doctors at Addenbrooke’s Hospital in Cambridge to train machine-learning algorithms to spot brain tumours in 3-D magnetic-resonance-imaging (MRI) scans.

By training the systems on MRI scans where tumours have been highlighted, the system can learn to pick out high-grade gliomas, tumours of the glial tissue within neurons in the brain, and to do so with a reasonable degree of accuracy.

Young says such approaches will help radiologists to more rapidly identify scans that need following up on, “taking a very human-intensive process of planning in the radiotherapy field and actually supercharging that radiologist so that they can do several of these maps a day”.

HIGHLIGHTING LUNG DISEASE – OPTELLUM

Composed of machine learning and clinical experts who met at Oxford University’s computer-vision lab, Optellum is able to analyze CT scans of nodules on the lung to help doctors spot signs of lung disease.

“Optellum do imaging of lung nodules of 1cm or less,” says Young.

“They have got up to around 98 per cent sensitivity and specificity to be able to tell you on your first CT scan whether that 1cm nodule is malignant or not. So, completely changing the way we manage those particular conditions.”

DETECTING EYE DISEASE – GOOGLE DEEPMIND

A well-publicized example, Google DeepMind has partnered with the likes of the Moorfields Eye Hospital to train an AI to scan images of patients’ retinas for signs of eye disease, first highlighting potential anomalies in scans, and then making treatment recommendations.

AI-GUIDED SURGERY

Young is bullish about the prospects for machine-learning systems that assist surgeons, via systems that help clinicians improve their surgical skills, highlighting the interactive surgical simulator provided by UK-based Touch Surgery.

“I have now seen the world’s first demonstration of live AI-guidance during a laparoscopic robotic operation,” he said.

“Their computer-vision systems have been developed so well they can map out, on screen in front of you, what the organs are and what instruments are coming into play. They can compare them [the surgeons doing the operation] to the world’s leading surgeons, telling them how they’re performing. Wow, truly amazing.”

DRONE DELIVERIES TO HOSPITALS

Inspired by work done by companies like Zipline and Matternet to deliver blood and pathology specimens between hospitals using drones, Young said the NHS was also looking into trialling drone deliveries between hospitals.

“We’ve been working with the UK Space Agency and Nesta through their Flying High report to look at how we might create safe corridors in cities – maybe along the Thames between Guys and St Thomas’ [Hospitals],” he said, adding that the medical director from Lincolnshire had asked him just last week whether it would be possible to use drones to solve a delivery issue between a couple of hospitals.

“I’m hoping that we’ll be starting with trials of that and working with partners in the next year of so.”

ROBOTIC STORAGE AND LOGISTICS

Looking further afield, Young believes the NHS could learn from the robotic technologies used by retail technology firm Ocado in its automated warehouses.

“This is the leading example of autonomous robots fulfilling a function. This was an industry that had no AI six years ago and they have focused on AI in very specific areas and are now world-leaders in this technology,” he said, highlighting the firm’s recent multi-billion dollar deal to build 20 fulfilment centres for US supermarket chain Kroger.

He speculated there could be a role for similar technologies “on NHS sites” for helping keep hospitals stocked with medical equipment and making sure equipment was delivered when needed.

“It’s the most exciting time in healthcare, I think it’s like when the printing press came along 500 years ago and democratized access to education,” said Young.

“Whether it’s artificial intelligence, advanced technology, robotics, social networks, personalization around the genome, all together those little steps we’ve been taking for a number of years now are combining to make one giant leap forward in healthcare.”

The NHS is due to receive a £20.5bn cash injection, which may help fund some of these advances, although when this amount is spread out over the five years to 2024, it will generally fall below the 4% annual increase needed just to maintain the existing level of service, according to Helen McKenna, senior fellow for policy and communications at The King’s Fund, an NHS-focused think tank.

“The long-run historic increase in NHS funding is around 4%, and growth hasn’t actually reached that level since 2010-11,” she said.

She has a more measured take on the technology’s potential to help the NHS at a difficult point in its history, where it’s struggling with “enormous workforce shortages”, “operational performance issues”, and widespread financial deficits in health trusts.

“It’s really important not to forget where we’re currently at,” she said.

“The health and care system needs to change and we all know that… but it’s important that we remember that AI and data analytics are ultimately tools to help us meet the big challenges of today.”

The post How machine learning, drones, and robotics will transform the NHS and healthcare. appeared first on Artificial Intelligence.

Avitas Systems, a GE subsidiary based in Boston, is now using drones and robots to automate the inspection of infrastructure such as pipelines, power lines, and transportation systems. The company is using off-the-shelf machine-learning technology from Nvidia (50 Smartest Companies 2017) to guide the checkups, and to automatically identify anomalies in the data collected.

The effort shows how low-cost drones and robotic systems—combined with rapid advances in machine learning—are making it possible to automate whole sectors of low-skill work. While there is plenty of worry about the automation of jobs in manufacturing and offices, routine security and safety inspections may be one of the first big areas to be undermined by advances in AI.

Drones have been used on some industrial sites for a while (see “New Boss on Construction Sites Is a Drone”), and various companies, such as Kespry, Flyability, and CyPhy, offer aerial systems for monitoring mines, inspecting wind turbines, and assessing building insurance claims. But the technology required to automate more of the process is now becoming accessible. Similar technology is also enabling robots to cruise autonomously through offices and malls looking for anomalous behavior .

Avitas uses drones, wheeled robots, and autonomous underwater vehicles to collect images required for inspection from oil refineries, gas pipelines, coolant towers, and other equipment. The company is using Nvidia’s DGX-1 system, a computer designed for a cutting-edge kind of machine learning, to guide these vehicles to the same spot, and to analyze the image data for possible defects.

Nvidia’s system employs deep learning, an approach that involves training a very large simulated neural network to recognize patterns in data, and which has proven especially good for image processing. It is possible, for example, to train a deep neural network to automatically identify faults in a power line by feeding in thousands of previous examples. In some cases, deep learning can perform image recognition more reliably than a person could.

Alex Tepper, the founder of Avitas, says the company’s customers spend hundreds of millions on manually inspecting equipment. This usually involves someone traveling to a remote location to examine it. A drone or robot can automatically collect images of the same spot many times over, perhaps making it easier to detect flaws that might otherwise go unnoticed. The approach can save a refinery, for example, about $1 million annually on inspections, the company estimates.

Advances in AI are also making it easier to teach robots to navigate to a location for themselves. This week, for instance, Neurala, a company that specializes in deep learning, launched a drone toolkit that can be used to train a vehicle to recognize or follow a particular object, and to avoid obstacles.

The post Drones and Robots Are Taking Over Industrial Inspection appeared first on Artificial Intelligence.

The farming industry is on the cusp of a so-called ‘technological revolution’. With drones, robots and intelligent monitoring systems now successfully being used in research and field trials, artificial intelligence, or machine learning, is set to revolutionise the future of farming as the next phase of Industrial Revolution in agriculture is on the horizon.

According to the UN Food and Agricultural Organisation (FAO), the global population is set to reach 9.2 billion by the year 2050. This means that the global agriculture sector is under more strain than ever with 2 billion more mouths to feed within the next 33 years. In such a scenario, a number of innovative agritech startups are coming up with increasingly accessible technology to transform the daily operations of the traditional family farms.

Vivek Rajkumar, the CEO-founder of Aibono, an agritech startup aimed holistically at turning ‘small’ to ‘smart’ in agriculture. The 30-something Social-Tech Entrepreneur who claims to be a third-generation farmer, quotes from the movie Martian, “There was a need to Science the hell out of Farming.” His Agritech Startup is transforming the model of agriculture from the ground up, attacking its inefficiencies by leveraging technologies, IoT, Crop Science and AI for farmers helping them grow profitable agri produce.

One of the biggest roadblocks to the growth of Indian agriculture is the low levels of yields. The predominant causes of low productivity are poor access to irrigation facilities; use of low quality seeds, low adoption of improved technology and lack of knowledge dissemination on improved agricultural practices. The challenge of small landholding size impacts diversification indices negatively. Technology and its access is a critical factor for diversified agriculture.

CXOToday: Can you share your views on how intelligent devices, such as robots and drones are enabling smart farming worldwide.

Rajkumar: There’s now worldwide pursuit towards Decision Agriculture powered by Data Sciences and Farm Analytics. We’re moving on from Precision Agriculture to Decision Agriculture. It’s an exciting time and we’d like to call it Agri 4.0 or the 4th Industrial Revolutionin farming. To set the context right, Precision Agriculture wave of the last decade is nothing short of impressive, from making inaccurate decisions on agro-chemical inputs without measurements, to technology around soil measurements and multispectral imaging of fields that enabled better decisions on agrochemicals, improving yields globally. Yet these remain been more congenial for large farms with economies of scale rather than smaller holdings.

While the last decade saw more broad-based decisions at farms, Industry 4.0, Intelligent Devices and Farm AI is rapidly enabling more granular decisions as well as in implementing these decisions in the farms. For example, “Variable Rate Fertigation” with a GPS enabled sprayer robot can vary the quantity of application per plant based on the ‘need’ at a hyperlocal level rather than treating 20,000 seedling in an acre of land as equals and applying the same quantity of fertilizer per plant; significantly increasing productivity, nutrient usage efficiency and an increased grower revenue. There are some good examples of start-ups from the West making farm robots that selectively removes weeds from Lettuce beds using image processing or apply herbicides.

The real potential in intelligent devices is in how it decentralizes scale to make efficient farming practices and higher productivity available for a farmer of smaller scale. Drones are replacing larger airplanes for crop dusting and satellites for imaging and Robots as a Service (RaaS) using machine learning and AI are replacing larger tractors. This era will be a game changer for a nation of small farms such as India, but for us, Robots and Intelligent devices will be ‘Part 2’ of Indian-ising Agriculture 4.0. The Part 1 is Information Services and AI.

CXOToday: How is it different in India? In a cost sensitive market like India, how can small entities afford these technologies?

Rajkumar: India holds the second largest agricultural land in the world, yet India’s agricultural landscape comprises 85% of small and marginal farmers operating on land holding of 2 acres of less. The average Indian small farmer, representing a 200 Million population, is therefore averse of capital or infrastructure spends, let alone intelligent or informed methods of farming. Lack of economies of scale had made modern technology, resources, experts or farm measurements unaffordable.

However, the 4th Revolution of Farming presents a wonderful opportunity to us, a nation of small farmers to bid adieu to poor yields and bad farm economics, that has been plaguing us for decades.

When it comes to India, we are entering an era of Internet Enabled Shared Services. For us, the answer in farming was probably never Capital. It is Services, Sharing & Data. We’ve got the fastest growth of affordable internet, smartphone adoption and the world’s largest population of youth (a 200 million) who can gather and interpret data from farms. I’d say the stage is set for Agri 4.0 and to turn our small farmers to smart farmers. Adoption of technology and the superior yields of Agri 4.0 will happen in two stages in India.Stage 1: At an Information level (now-2022): Shared Services & Equipment, Data Science and AI that helps small farmers make intelligent and informed decisions all the way from choice of crops to hedging risks to precise day-to-day agrochemical application for maximising yields and return of efforts.Stage 2: At a Hardware Level (2022-future): Farmers sharing smarter farm machinery and hardware with small form factor and higher degree of intelligence enabled by AI & Cloud, affordable for small holdings.

– CXOToday: Do you think farmers would be willing to trust such high-end tech solutions and researchers more than their age-old traditions?

Rajkumar: In our experience, for a farmer, seeing is believing. Better yields, better harvests, uniformly looking farm as opposed to sparsely spread out crops – are a few critical aspects that triggers a farmer’s attention. Despite age old traditional methods, a farmer today is more dependent on NPK fertilizers from the local store.

With respect to high-end tech solutions, Aibono provides Smart Farming Services, shared among Collectives of farmers that Aibono aggregates. Farmers share Sensors, Farm Managers, Data, Intelligence, Farm Experts, Tech Support, & Farm Equipment backed by AI and Data Analytics and in our Lab Farms, we demonstrate data led decisions. We earn the trust of a farmer and integrate ourselves around him and his community, which makes our Smart Farming Collectives both people and technology centric, and on the Cloud and on the Ground.

Also, the farmers are a sweet bunch – if they see results and get to like something, they spread the word our fast. Word of mouth is even faster than an app making the runs in an urban community. They are very results-driven and value science, agricultural expertise and modern methods.

– CXOToday: What was the idea behind Aibono? How exactly are you helping the farming agricultural sector in scaling up and remain sustainable?

Rajkumar: Aibono began in the niche area of providing Farm Management as-a-Service whereby, a farmer gets to outsource his entire measurement, production management and decision-making processes to a Service. We provide this service on a sharing basis deploying a shared Farm Manager along with shared instruments mapping the data onto cloud. Our centrally managed Data Science and Recommendation Engines enabled our by Data Scientists and agronomists give precise day-to-day interventions to farmers, enabling a 30-50% increase in yields.

Further, we evolved this model into incorporating what we call ‘Smart Farming Collectives™’ where groups of farmers collectively produce a uniform supply of a given mix of produce by sharing decisions and opportunities of choice of crops and collectively sharing resources. Thus they don’t compete amongst each other by producing the same crop, and the collective also enables the farmer to reach the end market. We’ve generated +1000 harvests, with 30-50% higher yields and consistently good incomes that come to the doorstep of the farmer. The sustainability of how we have interpreted Agri 4.0 for India, is really in the improved P&Ls of the farmer, our ally, enabled through our Smart Farming Collectives.

– CXOToday: At the moment, what are some of the biggest challenges in agricultural practices you are trying to solve with technology?

Rajkumar: One of the biggest roadblocks to the growth of Indian agriculture is the low levels of yields. The predominant causes of low productivity are poor access to irrigation facilities; use of low quality seeds, low adoption of improved technology and lack of knowledge dissemination on improved agricultural practices. The challenge of small landholding size impacts diversification indices negatively. All are triggered by economies of scale of a small farm. Farming yields in India can be as low as 20%-30% of Global benchmarks and our farming community has been cornered for decades due to lack of scale.

Technology and its access is a critical factor for diversified agriculture. With Internet Enabled Shares Services and our Smart Farming Collectives, we leverage Sharing & Aggregation to re-create economies of scale and increase yields by an order. We don’t stop there, we go the whole length to get the farmer a buyer and the produce a rightful place in the value chain, an enable a better income and return on effort that justifies his efforts. And, everybody wins.

– CXOToday: Can you give an instance where smart modern technologies such as big data, IoT and AI are being used to resolve some of these issues?

Rajkumar: Farm Analytics is one of the hottest drivers of the farming industry right now. The end consumer in agriculture is not always very aware of what trends have arrived now. Take this company called Solum Inc, an Iowa-based Agri-Data science company, whose soil sciences arm got acquired by weather & farming data analytics company called Climate Corporation, which again was acquired by Monsanto for $930 million, which in turn got acquired by Baeyer- The world’s largest agrochemical company- for $66 billion. All of this, in the last couple of years.

– CXOToday: What technologies are you offering farmers to maintain and scale crop diversity and intensive farming?

Rajkumar: An average vegetable farmer in India grow 4-6 crops cycle a year. Aibono’s farmers grow up to 24 crops per year with ERP Assisted Farming. Crop management for a small landholding farmer is quite confusing. In a Factory, ERP can enable small teams managing simultaneous batches, SKU variants and changeovers handle complex processes and reminders with ease. We provide Assisted ERP services for complex farm management processes, that not only enables our collectives of small farmers to push beyond 6 crops a year to 24 crops/year – even though it makes the farmer’s Crop Management and maintaining diversity 4x complex. The benefits however are significantly better portfolio risk management and more uniform earnings that outweigh the hassles.

– CXOToday: What do you see as the future of farming in terms of using smart technologies?

Rajkumar: India, as a country, has one of the largest Youth population, with a complex diversity and an infamous Jugaad mind that can adopt new tools and frugally solve problems and now we have taken the world by surprise in our ability to be the fastest adopters of smartphones and the internet. There are 400 million smartphone users in India – and aside just a fraction in cities, our use base is far beyond urban. We are after all the IT hub of the world.

The industry so far is depended heavily on scale of the user and therefore struggled to improve lives of small farmers in India. Agri 4.0 is perfect for an Evergreen Revolution in India, favouring small farmers and leveraging the IT & Smartphone Savvy Youth as a bridge between our farmers and cloud. Driven by internet, commoditised smartphones, smart sensors, AI and Cloud, Agri 4.0 is about Scale at the Cyberspace but Distributed in the physical space. Aibono as a company is marching towards a Self-Service model, where the farmers will be able to enrol for Shared services, ERP, shared equipment, farmer collective and intelligent market access and supply chain, where they will be enabled to sign-up and operate by themselves.

The farmer of the future will not have to go out of his way to access AI, Cloud, Farm Robots, Market Access or an Income that justifies his efforts. He will be served at his farm thanks to concepts like cyber physical share farming.

The post Artificial Intelligence Is Changing The Farms of the Future appeared first on Artificial Intelligence.