AI has changed the way we farm – and how!
Agriculture – an undoubted pillar of the global economy, and the lifeblood of a country such as India (which ranks second in the world for total agricultural output) – is rapidly becoming one of the primary fields benefitting directly from advancements in technology. Although mechanisation and adoption of new and more scientific farming methods have been prevalent in the world since time immemorial, a major part of agriculture still continues to be non-digitised and non-automated. The primary cause is a lack of technological prowess. This reality is, however, changing – driven by technologies, at the core of which lie artificial intelligence and machine learning.
The next Agricultural Revolution?
The first agricultural revolution of the modern world (and the second of the known world) was noted primarily in Britain between the mid-17th to the early-19th century – a period which saw a meteoric rise in the growth of crops owing to the use of techniques such as selective breeding, crop rotation and more prudent use of arable land. Since then, and until the 1920s, although agricultural productivity remained relatively high, it did often suffer from odd crop failures or famines.
There was, however, a dramatic shift in global agricultural standards owing to research advancements and a corresponding rise in investments in the 1930s. The Green Revolution – a set of research technology transfer initiatives between 1950 and the late 1960s – completely transformed the landscape of global agriculture. Production markedly increased all over the world, with cereal crops tripling in volume with only a 30% corresponding rise in arable land area. This increase was primarily due to genetic crop improvements, use of agrochemicals and mechanisation leading to productivity boosts.
Cut to 2020, and the next agricultural revolution is at hand. But this time, the driving force is Artificial Intelligence. AI and machine learning technologies have promoted automation within the industry, forcing a complete rethink of existing business models and establishing new product categories. A recent research report from Cambridge-based research firm IDTechEx, titled “Agricultural robots and AI-based machine vision represent the natural evolution of agricultural machines and tools”, believes that by 2040 the use of drones and agricultural robots will completely transform the way we farm.
The Drivers of the Evolution
The IDTechEx report comprehensively analyses all major emerging product types that are set to drive the oncoming revolution – “autonomous robots taking plant-specific precision action, intelligent vision-enabled robotic implements, diverse robotic fresh fruit harvesters, highly automated and autonomous tractors and high-power farm vehicles, drones, automatic milking, and so on”.
Autonomous Ultrapecision Robots: Small/mid-sized ultraprecision robots will use Convoluted Neural Networks (CNN)-based machine vision technologies that will not only allow robots to accurately identify individual plant parts, but also carry out health checks and tailored-processes required for individual attention. Stress will be laid on precision weeding, boosting yield, reducing the use of chemicals and better use of soil. Robotic tractor-pulled implements will be fully equipped with cameras and GPUs – and will essentially work as a computer vision system on wheels. Productivity will be crucial, with latest versions already capable of travelling at speeds of 12mph, processing images at 20fps at 2-inch resolution. These numbers are only set to become stronger.
Autonomous Spraying Vehicles: Autonomous water spraying has several benefits: (1) they can spray during very early mornings or very late nights to avoid evaporation; (2) can maintain a constant spray speed to achieve uniformity; (3) can relieve staffing pressure/costs and reduce exposure to legal claims for exposure to chemicals. The GPS-enabled robots will be designed accordingly and will be able to identify roadblocks as and when it comes across them. Advancements will still require moving from diesel to a more eco-friendly power source, but that improvement is already in the works.
Robotic Harvesters: Automatic harvesters and fruit pickers have been in the works for a long time now – and are now finally making the leap from laboratory prototypes to full-time working models. The report explains the development process by providing a broad outline: “The directions of product development have been the following: (1) improving the algorithm mainly using CNN techniques to cut down picking time; (2) extending the algorithms to recognize even more fruits so that the robot can be deployed across the entire calendar year; (3) designing novel end effectors to enable the gentle picking of the fruits; (4) selecting the right robotic arm (articulated vs delta) with the right performance-cost balance; (5) increasing the number of arms to boost productivity; (6) ruggedizing the system for the agricultural environment; and so on.”
With these algorithm-driven tools talking over the farms, agriculture as we know might be changing forever.
Acknowledgement:Agricultural Robots, Drones, and AI: 2020-2040: Technologies, Markets, and Players – reportsponsored by IDTechEx