.jpeg)
Precision agriculture with features of predictive modeling, real-time data capture, use of digital farm data, Internet of Things (IoT), sensors, etc., has been seen as a step forward in enhancing the efficiency of farm processes.
However, this concept is restricted to farming while the present paradigm needs to encompass the whole food system.
Digital agriculture generally referred to as Agriculture 4.0, is an evolution of precision agriculture and is set to create a new paradigm in India in the complete food systems cycle.
It involves the use of sensors, IoT, big data, artificial intelligence (AI), satellites, and robotics. Industry and service sectors which contribute most to the national gross domestic product (GDP) have progressed well in the use of digital technologies, while agriculture has lagged behind.
Agriculture could learn from the other sectors and directly focus on ‘Good data’ in place of ‘Big Data’. In this paper, a very general appraisal is given to sensitize people from farming, agricultural sciences, industry, policy makers and general public on the need and role of digital technologies in Indian agriculture.
What makes digital technologies paradigm backed in INDIA?
The first 3 paradigms of agriculture, viz., farm mechanization, chemical agriculture and crop and animal genetic advances are followed by data-driven agriculture, touted as the fourth.
Agriculture 4.0 is an evolution of precision agriculture and refers to the use of internet of things (IoT), big data, artificial intelligence (AI) and robotics to extend, speed up, and increase the efficiency of activities that affect the entire food production chain.
A technological revolution in farming led by advances in robotics and sensing technologies looks set to disrupt modern practice.
Farming is the next frontier for using artificial intelligence (AI) to efficiently solve complex problems.
The benefits of digital agriculture are food security; quality of soil, air and water; better economic returns; and efficiency of crop and animal production and quality of life.
The digital ecosystem in agriculture encompasses diverse players - government, research, industry, markets, social and ecological domains.
New and emerging technologies such as the internet of things (IoT), drones, mobility, cloud computing, big data, remote sensing, artificial intelligence (AI), machine learning (ML), image analytics and processing, block chain, and agribots are poised to transform traditional agriculture into data-driven precision farming to generate sustainable profits.
Digital agriculture can be defined as the modification or addition of processes within food systems that are substantially enabled or enhanced by digital technology. Digital agriculture is the seamless integration of digital technologies into crop and livestock management and other processes in agriculture.
These data are acquired from different sources such as field surveys, weather stations, historical datasets, geospatial data, satellite imageries, sensor-crops, animals, soil, farm equipment mounted and GPS-based field maps.
Since the data retrieved from weather stations may not be accurate and may not represent prevailing crop/animal microclimatic conditions, the plant/animal growth algorithms developed have to be reset and the recalibrated advisories at farm level need to be developed.
Precision farming at small-farm holdings needs solutions at field level. New technologies in field data collection with more frequent data and minimum manual errors include IoT - based sensing systems that monitor precisely crop parameters, thermal imaging and environmental conditions.
These sensing systems are connected to wireless sensor network (WSN) and are collected at a central location. The modern networks are bi-directional, both collecting data from distributed sensors and enabling control of sensor activity. The IoT connected with WSNs will provide a basic framework for high-density data collection, for example, microclimate (temperature, relative humidity, rainfall, wind speed, solar radiation, etc.), soil and plant parameters and help to solve critical issues about the crop-weather-soil continuum level (within field) crop monitoring is essential to precision farming, which requires frequent crop growth information at high-spatial resolution for mid-season crop management.
Recently, deep learning-based (DL) algorithms of AI are getting attention in multi-scale data analysis for agricultural application.
Those techniques are based on artificial neural network (ANN)/convolution neural network (CNN) and these introduce new approaches to the prediction of problems by adding the capability to learn possibly noisy and non-linear relationships with arbitrarily defined, but fixed number of inputs and outputs supporting multivariate and multi-step forecasting.
Integrated modelling and data analytics include image analytics, identification of crop phenology and yield estimation and crop health monitoring and prediction.
The algorithms developed may not be universally applicable due to varying climatic conditions, crop types, and needs of farmers and moreover farmers may trust human agronomist more over AI-based advisories.
Unmanned Aerial Vehicles (UAVs) equipped with different imaging sensors such as RGB (Red Green Blue), Multispectral (MSI), Hyperspectral (HSI), Lidar are flown over crop fields to acquire the data, which is further, analyzed using AI/ML techniques to study the phenotypic traits, nutrient analysis, water, disease stress detection, biomass and crop yield estimation.
Integration of unmanned aerial vehicle with satellite based remote sensing images has been suggested.
A recent start-up-promoted by a former ISRO scientist promises to make synthetic aperture radars (SAR) which can be fitted to drones and flown at low altitudes to get more accurate images (12 March, 2022 Former ISRO scientist’s startup to make low-altitude, high-resolution radars | Cities News,The Indian Express ) https://indianexpress.com/article/cities/ahmedabad/former-isro-scientists-startup-make-low-altitude-high-resolution-radars-7815506/which have potential to solve on-farm problems more accurately. Image analytics and AI need to be tuned to enable precision application of agro-chemicals by drones in order to economies inputs and reduce environmental footprints.
During kharif 2021, Telangana IT Department conducted an artificial intelligence-based pilot project for pest management in cotton in six districts where using pest incidence imageries helped timely pest control https://telanganatoday.com/telangana-ai-helps-yield-bumper-cotton. If further data analytics are used, such gains could be much higher and focused.
Field level agronomic solutions alone will fuel digital services to farmers. In order to meet the emerging demands of digital agriculture, scientists should focus on research on development of process-based algorithms, constant up-gradation of ML (machine learning) and DL (deep learning) based algorithms powered by dynamical feedback mechanisms.
The mind set of one-time technology development has to be replaced by dynamical agronomy solutions. This is possible only if agricultural scientists collaborate and work with diverse fields such as IT professionals, earth system scientists, socio-economics professionals, and more importantly business and management professionals.
Educating the agricultural workforce with the skills to manage and harness the power of these new digital tools will be crucial to achieve mass adoption.
BY KHAS.
Comments