Given the significant advances made by Artificial Intelligence, it’s time India makes good use of this technology to encourage higher farm yields and weed out the many problems that bedevil this sector
Artificial intelligence (AI) is much closer to our lives than ever before. Be it Apple’s Siri or Google’s Alexa, every intelligent assistant relies on speech recognition software whose backbone is AI. But for its untapped potential, human beings are turning to AI not only to play their favourite song from the playlist, but also to safeguard the environment, win over the climate change challenge and increase agricultural productivity. AI-driven technology holds immense potential. The best-laid plans, emanating from the elite minds in the scientific world, have neither been able to stave off worsening environmental conditions, nor could they ensure better state of affairs for our agricultural sector.
AI is fundamentally the simulation of human intelligence by computer systems. It has been around for quite some time, in fact since 1950s, but during those times, it was more of a concept than an applicable technology. Thanks to rapid technological progress, things have changed now. Today, AI holds the potential to think, learn and act in response to its immediate environment, according to its programmed objectives. These smart attributes can prove to be a game-changer if AI is applied effectively on a large-scale basis for environmental protection and climate change.
The environmental situation, as it stands today, looks abysmal. The planet continues to warm up and extreme weather events are multiplying rapidly. Environment conditions and worsening and climate change circumstances are assuming severe proportions with each passing year. In 2016, there were 772 weather and disaster events, triple the number that occurred in 1980. Due to such events, around 20 per cent of the animal species are under the threat of extinction. This number is only expected to rise to 50 per cent by 2100. The combined scientific prowess of the world and the unified resolve to limit spiralling global temperatures have not helped. In fact, even if all signatory countries stand by their pledge at the Paris climate deal, by 2100 it’s likely that average global temperatures will still be three degrees Celsius higher than in pre-industrial times.
AI can, however, change the environment for better, quite literally. Starting with weather and pollution forecasting options, AI can discard those models whose prediction rates are abysmal. This can help achieve accuracy. Similarly, a smart AI software can simulate urban heat waves and predict which industries should be shut down in order to lower the temperatures. They can even suggest the locations to plant trees so as to reduce the urban heat island effect. These are some of the many advantages that AI technology holds. India must, therefore, explore the possibility of making this technology a part of the daily environmental protection measure.
As urban India endures the debilitating impacts of climate change, rural counterparts, especially farmers, fare no better. They are faced with an ever-increasing challenge to maintain farm productivity in the midst of rising agricultural debts and unpredictable weather patterns. These conditions in the agricultural sector have, on the one hand, caused a spate of farmer suicides due to rising debts, and on the other, these trying situations have led to a steady drop in agricultural productivity, leading to a possible threat to food security in the near future.
Even though in a nascent stage, AI is slowly but surely making its presence felt in the agricultural sector in our country. For instance, digital farming practices, like cloud technology and business intelligence, have helped groundnut farmers in south India get 30 per cent higher yields per hectare. Using AI, they could get information related to preparing their land, applying fertiliser and choosing sowing dates. AI even helped researchers achieve 89 to 99 per cent accuracy in identifying tropical cyclones, weather fronts and atmospheric rivers, which are often hard for humans to identify on their own. By improving predictability of weather-related events and intensity as well as the duration of weather spells, it has now become easier for farmers to plan their crop cycles, which is a godsend in times of climate change induced by weather ambiguities.
AI can alter the future of Indian agriculture. It can weed out the problems that have bedevilled this sector. The power of technology can help take drudgery out of agriculture and usher in an era of e-agriculture or smart farming. AI technology can take over planting, maintaining, harvesting crops and detect certain diseases in plants and, thus, help save money, energy, labour and resources. The smart software driving AI can build programme objectives in such a way that crop diseases and pest damages are recognised. Measures are automatically taken to protect the crop. In a trial simulation, researchers used TensorFlow, Google’s open source library, to build an archive of 2,756 images of cassava leaves from plants in Tanzania. The AI-run crop protection software, that relied on this database, was able to identify a disease with 98 per cent accuracy. This shows the potential of AI.
Agriculture is slowly becoming digital with AI showing promising potential in three major categories: Agricultural robotics, soil and crop monitoring and predictive analytics. Farmers are already using sensors and soil sampling to gather data which is stored on farm management systems that allow better processing and analysis. This data, along with specific algorithms, can be combined with weather information sourced from satellites to create customised AI software for different agricultural regions in India.
The global use of AI for agriculture is quite impressive. For instance, the US-based Blue River technology, a John Deere subsidiary, has evolved machine learning to a stage where sensors detect weeds, their type and right herbicides to apply within the right buffer around the plant. Cameras and sensors use machine learning that can capture images. Besides, the machines can be taught about different weeds. Even then, the right herbicides are sprayed precisely as per encroachment area.
Blue River Technology has further developed a robot called ‘see and spray’, which reportedly leverages computer vision to monitor and precisely spray weeds on cotton plants. Precision spraying can help prevent herbicide resistance. Similarly, Harvest CROO Robotics has developed a robot to help strawberry farmers pick and pack their crops. The lack of labourers has reportedly led to millions of dollars of revenue losses in key farming regions such as California and Arizona. Apart from this, 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 diseases, a smart phone collects image, which is matched with a server image and then a diagnosis of the plant health is provided. In this way, it uses AI and machine learning to solve threatening plant diseases.
Given the adoption of emerging technologies, AI is expensive and out of reach for the common man, especially for debt-ridden farmers. But this is where the opportunity lies — India has the advantage of scale and scientific resources that can rival the best of the world’s research institutions. These strong points provide India with enough raw materials to build a robust and comprehensive AI software that can provide one-time solution for all issues being faced by farmers in the effort to produce a bumper harvest.
In this quest of becoming pioneers of largescale AI application in agriculture, India must look towards the Indian Space Research Organisation to provide the necessary satellite data, as exhaustive information is the backbone of any successful AI software. This information, processed through latest computers, will give AI the edge it needs and ensure fast data processing. This will in turn enable AI to quickly discern patterns that humans cannot, make predictions more efficiently and recommend better policies — all in record time. But today’s AI still has a lot of catching up to do when compared to humans. For instance, it takes 83,000 processors, 40 minutes to compute what one per cent of the human brain can calculate in one second. This needs to improve drastically for AI to be a game changer.
(The writer is an environmental journalist)
