Home » Artificial Intelligence-driven Technologies in Agriculture

Artificial Intelligence-driven Technologies in Agriculture

Advertisements

According to United Nation Food and Agriculture Organization, the global population will increase by two billion by 2050, but only 4% additional land will come under cultivation by then. Farmers are facing with an ever-increasing challenge to maintain farm-productivity in the midst of rising agricultural debts and unpredictable weather patterns. These conditions in Indian agricultural sectors have caused a spate of farmers’ suicides due to rising debts.  Agricultural productivity has dropped, leading to a possible threat to food security in near future. While large scale research is still in progress and some applications are already available in the market but the agricultural sectors is still highly under-served. When it comes to handling realistic challenges faces by farmers and using autonomous decision making and prediction solutions to solved them, farming is still at a nascent stage. Hence, use of latest technological solution to make farming more efficient, remains one of the greatest imperatives. The future of farming depends largely on adoption of cognition solutions.

Artificial Intelligence (AI) is one of the key areas of research in computer science. AI is becoming pervasive very rapidly due to its robust applicability in several problems which cannot be solved well by traditional computing structure and human. In such an area of extremely importance in agriculture which faces several challenges from sowing to harvest, AI-driven techniques and tools can be employed to encounter those challenges in agriculture. The major issues of agriculture are inadequate application of fertilizer and nutrients, improper drainage and irrigation, weed control, disease and pest infestation, yield prediction.

Background of Artificial Intelligence

The first AI program, The Logic Theorist, was designed by Newell and Simon in 1955. The term Artificial Intelligence was coined by John McCarthy who is regarded as the father of AI. During the early 1980s and 1990s, the rule based expert system was extremely used, but from 1990 onward, artificial neural network (ANN) model and fuzzy inference systems have taken the dominant role. There are three major AI techniques such as expert system, artificial neural network and fuzzy system. In recent years an uprising use of hybrid system such as neuro-fuzzy or image processing coupled with artificial neural network are being used. AI is the simulation of human intelligence processes by machines and computer systems. Major sub-fields of AI include neural networks, machine learning, expert systems, speech processing, natural language processing, robotics and planning. There are several types of AI such as artificial general intelligence (AGI), artificial narrow intelligence (ANI), artificial superhuman intelligence (ASI). It moves toward more automated and more accurate systems that act on real-time.

AI technique is a manner to organize and use the knowledge efficiently in such a way that it could be perceivable by the people who provide it. It should be easily modifiable to correct errors and used in many situations. AI is becoming the important part of our daily life. Our life is changed by AI because this technology is used in a wide area of day to day services such as speech recognition, machine vision. Internet of Things (IOT) includes tools such as robotics, drones, GPS & remote sensing technologies and computer imaging. Various low-cost sensors on field and in space helps in determining soil conditions, groundwater levels, Chlorophyll Index highlights crop stress in time, ensuring harvest predictivity at each crop stages. The adoption of IoT devices in agriculture is on the boom, which speaks a testimony in implementing technology-based data-driven agricultural practices.

Artificial Intelligence in agriculture is the creation and study of computers and software’s capable of intelligent behavior which helps in creating ‘Self Learning Algorithms & Capabilities’, leading to automation on-ground agriculture practices. Farm activities such as field sowing, ploughing, fertilizer application, insecticide spray, harvesting, weeding and post-harvest land-replenishing can be carried out by the applications and processes developed around AI. AI comes as a great boon to the agricultural sector which is slowly but surely making its presence in agricultural sector.

AIMy

Crop Management in Agriculture

General crop management system provides an interface for overall management of crops covering each aspect of farming. Issues pertaining to soil and irrigation management are very vital in agriculture. Improper irrigation and soil management lead to crop loss and degradation quality. Application of herbicides has a direct implication on human health and environment as well. Modern AI methods are being applied to minimize the herbicide application through proper and precise weed management. Insect pest infestation is one of the most alarming problems in agriculture that lead to heavy economic losses. Researchers have tried to mitigate this menace by development of computerized system that could identify the active pests and suggest control measures. Crop diseases are also a matter of grave concern to a farmer. Significant expertise and experience is required so as to detect an ailing plant and to take necessary steps for recovery. Computer-aided systems are being used globally to diagnose the disease and to suggest control measures.

Agriculture is a dynamic domain where solutions cannot be generalized to suggest a common solution. AI techniques have enabled us to capture the intricate details of each situation and provide a solution that is best fit for that particular problem. AI can be employed for agricultural product monitoring and storage control.  Storage, drying and grading of the harvested crops are important aspects of agriculture.  Hence, AI can be employed in addressing various food monitoring and quality control mechanisms. The crop yield prediction is very beneficial for marketing strategies and crop cost estimation.

 Artificial Intelligence-based Technologies in Indian Agriculture

Artificial Intelligence can alter the future of Indian agriculture which is primarily dependent on unpredictable climatic conditions. AI is being applied in every nook and corner of human endeavours but agriculture has fallen behind its adoption. AI is basically the simulation of human intelligence by computer systems. AI has potential to think, learn and act in response to its immediate environment according to its programmed objectives. AI-driven technologies can take over planting, irrigation, maintaining, harvesting crops and detect certain disease in plants, leading to save money, energy, labour and resources. Technologies such as satellite image analysis, machine learning and cloud computing are revolutionizing agriculture. These technologies are helping farmers through the timely delivery of weather information, prices and automated calls to protect their crops from pests and weeds. These predictions are primarily based on weather conditions and other factors. Statistically predictions and Machine Learning (ML) play important roles in these processes.

A central task force on Artificial Intelligence has suggested creating a National Artificial Intelligence Mission (N-AIM) that will serve as a nodal agency for coordinating AI related activities in the country. Professor V Kamakoti, a Professor of IIT, Madras is the chairman of the task force and members include those from government, academia and the private sector. According to the report of the task force, set up by the Union Commerce Ministry, AI is the science and engineering of making intelligent machines, especially intelligent computer programmes and it should be seen as a scalable problem solver in India rather than only as a booster of economic growth. The task force has identified ten domains in India including Agriculture and Food Processing for implementation.

Application of AI in agriculture broadly comprises soil and crop monitoring which involve in use of IoT and sensors technologies to monitor soil and crop health, predictive farming analytics which involve in using machine learning to help farmers plan for sowing and reaping calendar, and supply chain efficiencies which involves  in using analytics to optimize the supply chain.

Internet of Things (IOT) includes tools such as robotics, drones, GPS & remote sensing technologies and computer imaging. Various low-cost sensors on field and in space helps in determining soil conditions, groundwater levels, chlorophyll index highlights crop stress in time, ensuring harvest predictivity at each crop stages. The adoption of IoT devices in agriculture is on the boom, which speaks a testimony in implementing technology-based data-driven agricultural practices. AI-based technologies applications in agriculture of these categories are used by various companies.

Companies/Startups using AI-based Technologies in Agriculture

Microsoft has begun empowering small-holder farmers in India to increase their income through greater price and higher crop yield. Microsoft is working with 175 farmers in Andhra Pradesh, India to provide advisory services for sowing, land preparation, fertilizer applications. International Crops Institute for the Semi-Arid Tropics (ICRISAT) working with Microsoft launched an AI-based Sowing App which sends advice to the farmers about the date to sow and other advice. Moisture Adequacy Index (MAI) is used to calculate the crop-sowing period. MAI is the standardized measure used for assessing the degree of adequacy of rainfall and soil moisture to meet the potential water requirement of crops. The App sends sowing advisories to participating farmers on the optimal date to sow crop. There is no need for farmers to install any sensor or device in their fields but only need a phone receiving a text message. This initiative has already resulted in 30% higher yield per hectare on an average compared to last year. Microsoft has also collaborated with United Phosphorous (UPL), India’s largest producer of agrochemicals, to create the Pest Risk Prediction App based on AI and ML to forecast in advance the risk of pest attach. Farmers across the Indian states of Karnataka, Andhra Pradesh, Maharashtra and Madhya Pradesh receive automated text message before sowing the seeds and voice calls for the alert of pest attack.

CropIn is a Bengaluru-based start-up company using AI to maximize per-acre value and claims to be an intelligent and self-evolving system delivering future-ready farming solutions to the agricultural sector. Basically, the CropIn uses proprietary machine learning algorithm built on satellite and weather data to give insight at plot and region level. The company used to record arm data manually from 2500 plus potato plots spread across an area of 5200 plus acres. All plots were geo-tagged to find the actual plot area and the ‘smartfarm’ solution helped in remote sensing and weather advisory, scheduling and monitoring farm activities for complete traceability, educating farmers on adoption of right package of practices and inputs, monitoring heath and harvest estimation and alerts on pest and diseases. The CropIn uses AI-driven technologies to help clients analyse and interpret data to derive real-time actionable insights on standing crop and its agri-business intelligence solution known as SmartRisk “leverages agri-alternate data and provides risk mitigation and forecasting for effective credit risk assessment and loan recovery assistance.

Intello Lab is a Bengaluru-based startup by IIT-Bombay alumnus Milan Sharma in May 2016. The company uses deep learning algorithms for image analysis and claims to provide image recognition technology which can recognize objects, faces, flora, fauna and tag them in any image. This new generation of intelligent image based applications provides insights on the crops’ health during the growing season and its final harvested quality by click of photograph. Farmers can click on image of their crop and use their solution to understand the diseases, pests and weeds growing in their farms. The solution uses DL and image processing models to identify any crop diseases or pest infestation in the crops and also provides recommendations on how that disease can be cured and prevented from increasing further.

Gobasco is a company co-founded by Vedant Katyar, an engineering graduate from premier Indian technology institute BITS Pilani and Abhishek Sharma, CTO, is a doctorate in AI from the University of Maryland, USA. This company based in Uttar Pradesh,India uses AI and related technologies in the various stages of the agri-supply chain. Real-time data analysis on multiple data-streams along with crowd-sourced from producer/buyer marketplaces and transporters feeds their automatic transaction discovery algorithms to obtain high-margin transactions. Crowd-sourced data, algorithms and analytics overcome the credit default problem, the most challenging problem of current supply-chain, to ensure a very low risk operation. Agri-Mapping involving a real-time agri map of commodities is obtained by deep-learning based satellite image analysis and crowd-sourced information fusion. For creating an international agri-commodity standard, computer vision and AI-based automatic grading and sorting of vegetables and fruits is done for quality maintenance and reliable trading across country boundaries.

Gramophone is a based in the Madhya Pradesh, India and Tauseef Khan is the co-founder of the company. This company uses AI and ML for image recognition technology in tandem with proprietary database.  Temperature and humidity data is used to guide and help farmers with timely information and right kind of inputs to achieve better yields. Pathology/entomology data is used to predict pest and disease risks. Gramphone is also used to forecast commodity prices for better price realization.

Jivabhumi co-founded by TS Srivatsa, is based on AI engine which leverage the comprehensive aggregation of data at various points in supply chain.  The platform captures comprehensive information about the commodities regarding growing information, pre- and post-harvest, transportation, warehousing to generate a digital identity for a physical commodity and build traceability to prove provenance and movement of commodities from farm to table. This ‘Footprint’ is a produce aggregation and food traceability solution which provides e-marketplaces services and implements traceability.

Aibono is a Bengaluru-based company started and founded by Vivek Rajkumar, a graduate from IIT-Madras in 2014. Aibono, Ai standing for Artificial Intelligence and bono meaning for the public good, provide real-time precision agriculture services to farmers.  In India, a majority of the farmers have small holding of 1.5-2 acres, are not in position to use technology or bring in experts to help them. Aibono uses technology based on internet, broadband connectivity and smart phones and uses sensors on the field to collect soil data, leaf coloration or other images and upload it on the platform. The data is analyzed to recommend to the farmer what need to be done on the day-to-day basis. The company  set up a model farm in Nilgiris, Tamil Nadu providing precision agriculture services to fruits and vegetables farmers. It tracks a large number of farm variables, uses data sciences to analyze that information and provides inputs to farmers. The farmers operating on Aibono’s platform have seen their yields go up 1.8-2 times.

The Berlin-based agricultural tech startup ‘PEAT’ has developed the Plantix app that identifies potential defects and nutrient deficiencies in the soil. This application uses images to detect plant disease; a smart phone collects image which is match with server image and then a diagnosis of the plant health is generated. Thus AI and ML are used to solve threatening plant disease.

Blue River Technology, a US-based company, has developed a robot known as “See and Spray” which leverages computer vision to monitor and precisely spray weeds on cotton plants. Precision spraying can help to prevent herbicide resistance. The ability to control weeds is a top priority for farmers and an ongoing challenge because of herbicide resistance problem. About 250 species of weeds have become resistance to herbicides.

Harvest CROO Robotics, a US-based company has developed a robot to help strawberry farmers pick and pack their crops and claims that its robot can harvest eight acres in a single day and replace thirty human laborers. Lack of laborers has reportedly led to millions of dollars of revenue losses in key farming regions such as California and Arizona.  Automation is also emerging in an effort to help address challenges in the labor force. The industry is projected to experience a six percent decline in agricultural workers from 2014 to 2024.

SkySquirrel Technologies is one of the companies bring drone technology to vineyards. The company uses algorithms to integrate and analyze and the captured images and data to provide a detailed report on the health of the vineyard. The company aims to help farmers to improve their crops yield and to reduce costs. Users pre-programm the drone’s route and once deployed the device will leverage computer vision to record images which will used for analysis and claims that its technology can scan a 50 acres in 24 minutes and provides data analysis with 95 per cent accuracy.

A Colorado-based company, ‘aWhere’, uses machine learning algorithms in connection with satellites to predict weather, analyze crop sustainability and evaluate farms for the presence of pest and diseases. The company claims to specialize in providing a high quality of data that is continuously updated at a rapid rate. Data sources include temperature, precipitation, wind speed, and solar radiation.  It also provide its users (farmers, farm consultants, researchers) with access to cover a billion points of agronomic data on a daily basis.

FarmShots, a company based in Raleigh, North Carolina US, is a startup focused on analyzing agricultural data derived from images captured by satellites and drone to detect pest and disease infestations and nutrition status on farms. The company claims that its software can inform users exactly where fertilizer is needed and can reduce the amount of fertilizer used by nearly 40 per cent.

California-based ‘Trace Genomics’ provides soil analysis services to farmers. After submitting a sample of their soil to the company, users reported receive an in-depth summary of their soils contents. Services are provided in packages which include a pathogen screening focused on fungi and bacteria.  The users can get information about a comprehensive microbial evaluation.  The emphasis is on preventing defective crops and optimizing the potential for healthy crop production.

SatSure, a London-based startup, uses satellite data and ML techniques to assess imageries of farms and predict monetary prospect of their future yield.  The company helps insurers and financers in deciding the value of agricultural land.  ‘Earth Food’ and ‘V Drone Agro’ are other similar companies which use AI to assess soil conations over the cloud and provide services to help farmers so that they can get maximum yield returns.

Gamaya is Switzerland-based company which offer a drone-mounted hyperspectral imaging camera and claims combines remote sensing, machine leaning and crop science technologies. The camera measure the light reflected by plants within visible and infrared light spectrum. The plants with different physiologies and characteristics reflect light differently. This pattern changes as the plant grow and is affected by stressors. Gamaya’s technology is capable of mapping and distinguishing the weeds from plants. It is able to identify other plant stresses such as disease, malnutrition, and other chemical inputs in the soil.

Neurala has developed the Neurala Brain, a deep learning application which the company requires less training, less data storage and less computing resources. This application may be implemented in drones, camera, smartphone. Training the algorithms uses the company’s Brain Builder data processing tool which enables users to upload and label their own sets. To start the training process, the user much upload to the device about eight images per subject. Neural’s algorithms will learn in 25 seconds. This is much faster than traditional deep neural networks which take more than 15 hours to be trained on a server and require 36 images per object.

Iris Automation developed the Iris Collision Avoiding Technology for Commercial drone, an application that allows drones to observe and interpret its surrounding and moving aircraft to avoid collision. This application is also suitable for use in agriculture where the drone application is capable of assisting farmers in surveying crops, controlling pests, planting seeds while interacting with other drones safely. The system’s computer vision gives it the ability to see obstacles, aircraft, and other potential dangers for a safe and reliable flight as it captures images of its surroundings during flight. Once the images are captured, the camera’s deep learning algorithms process he data by finding similar images in its database to recognize the object. Recognizing the objects allows the drone to then know where to fly.

SenseFly offers the ‘Ag360’ computer vision drone which capture infrared images of fields to help farm owners monitor crops at different stages of growth and assess the condition of the soil. This could enable farmers to keep track of plant health and determine the amount of fertilizer needed to be applied to avoid wastage.  During the flight, the drone captures the imaging data of the field while ‘eMoton’ then directly uploads these images to cloud services. The ‘Pix4Dfield’ image processing application generates aerial maps of fields for crop analysis. Its algorithms translate the image data to create maps of the field by finding matching images within its database to recognize the condition of the plants and soil. Farmers, agronomists, soil scientists and breeders provide the inputs to train this application. The maps enable farmers to determine soil characteristics such as moisture and temperature, and provide guidance to improve crop growth and production.

Challenges and Future Prospects

Although the use of AI is promising but there are challenges when it comes to agriculture. The development of AI algorithms in agriculture is one of the most challenging jobs because of large amount of data requirement to train the algorithms.  Other challenges include non-availability of data from remote areas and availability of limited data once per year during the growing season only.  Prediction in agriculture is still complex and elusive. Marketing is the critical factor in driving framers from poverty to prosperity.  NITI Aayog is working in this direction to solve these challenges by digitalization of agriculture. Digital India is reaching in villages of India for leveraging AI techniques in agriculture. NITI Aayog’s Statement of Intent (SoI) to develop and deploy AI to provide real-time advisory to farmers should be extended all over the country. Even though in a nascent stage, AI is slowly but surely making its presence felt in the agricultural sector. Global use of AI for agriculture is quite impressive.

Agricultural robotics, soil and crop monitoring and predictive analytics are the major categories where AI based technology are used. Agriculture is slowly becoming digital with AI and using sensors, drones, robotics, satellite images, weather information, soil  testing data, soil moisture and so on. Drones in agriculture are used for soil and field analysis, planting, crop spraying, crop monitoring, irrigation, and health assessment.  AI-driven drones can be used for more advanced data-gathering tools. Technologies such as Artificial Intelligence, cloud machine learning, satellite imagery and advanced analytics will empower the small-holder farmers to increase their income through higher crop yield and greater price control.

Conclusions

Agriculture is slowly becoming digital with AI showing promising potential in categories such as soil and crop monitoring, predictive analytics and robotics. Sensors and soil sampling to gather data are already in use to store farm management systems information for processing and analysis. This data, algorithms, weather information and images from satellites are used to create AI based software for different agricultural regions in India. An open source platform would make the solutions more affordable, resulting in rapid adoption and higher penetration among the farmers. Though currently application of AI in agriculture is in a nascent stage but with time & capital investment, farm mapping, observation, predictability and on ground farm operations will be automated, leading to increase in efficiency and reduction in production cost and minimizing environmental impact. AI-based technology can take over planting, maintaining, harvesting crops, grading fruits and vegetables and detect certain disease in plants. AI-powered solutions will not only enable farmers to do more with less, it will also improve quality and ensures faster go-to-market for crops. AI-based technology can solve the problems that are present in agricultural sector leading to take agriculture in the era of e-agriculture or smart farming. Hence, this smart farming using AI-driven technologies can alter the future of Indian agriculture and can bring a paradigm shift in how we see farming today.

Advertisements
%d bloggers like this: