For thousands of years, farmers have been looking up to the clouds to try to decipher what’s in store for their crops this season. Today, a new cloud is on the horizon. More and more farmers are turning to big data and precision farming, in order to maximize the potential of their land and resources as well as the quality of their produce, while minimizing the impact on the environment.
Why on earth?
As the human population grows and resources become more and more scarce, specifically water and arable land, innovative solutions are required in order to meet the increasing demand. According to the United Nations’ World Population Prospects report, by the year 2050 the earth’s population will grow to almost ten billion people. In addition to the swelling population, there is very little arable land left on the planet to grow more crops. Of course, deforestation is unacceptable, because of the toll it would take on the environment, and at the same time, potential farmland is being lost to desertification, urbanization and other phenomena which might lead to an overall decrease in arable land in the future. Water is also a limited resource, and has become less predictable and harder to efficiently collect given the recent climate changes. At the same time, consumption in cities continues to increase, pushing prices up for farmers and making irrigation a major expenditure.
Besides the growth in population, there is expected to be a shift in many places, especially India and China, towards a growing middle class population. This, in turn, is anticipated to drive demand even higher than the actual growth in population, and leads to estimates of approximately 70% in overall demand for food in the next few decades.
All these facts and projections paint a pretty bleak picture of the future. Fortunately, many farmers, scientists, and technology experts are working tirelessly to face these challenges. Let’s explore some of the solutions, which are currently in use or in development, that aim to prevent any shortage in food while keeping the environment intact at the same time.
Enter Precision Agriculture
In an article from 2012, Dr. Jim Budzynzki describes the evolution of farming in the 20th century. Ag1.0, according to this description, was labor intensive, low productivity, traditional farming. Ag2.0 was the dramatic shift in the industry that mostly took place in the second half of the twentieth century. This shift included the introduction of synthetic fertilizers and pesticides, significantly driving up efficiency and reducing differentiation, effectively cutting costs. On the economic front, this went together with vaster global marketing and the industrialization and growth in scale of farms. All these changes were accompanied by very low environmental awareness, leading to some upsetting consequences. Ag3.0, he suggests, is the new paradigm which will bring with it a huge surge in the use of technology to accurately manage farms, while shifting focus from efficiency to profitability. A phrase he suggests will be in use is “No molecule wasted”.
It seems that’s a pretty good description of where the agriculture industry is headed. In today’s technologically pioneering farms, autonomous combines are already harvesting crops using extremely precise Global Positioning Systems (GPS), accurate down to the centimeter, while using special sensors on the ground to perform grid sampling of the fields. Drones fly overhead, mapping the field and collecting millions of different data points, and sending them to the cloud to be processed. Meanwhile, the farmer looks at his tablet, immediately identifying areas that require special attention. The collected information is then fed back into the tractors and choppers which automatically make corrections for amounts of pest control substance. This scenario is still quite rare, but more and more farmers are adapting to this revolution in agriculture brought on by information technology.
According to Nasa’s Earth Observatory, measurements in visible, near-infrared, thermal infrared, and microwave wavelengths of light can indicate when crops are under stress. Identifying these crops in time, and correcting the application of nutrients or fine-tuning the irrigation process can save these crops and improve their quality. By applying sensors with the relevant vision capabilities (drones, automated tractors), farmers can now determine the precise care that each plant needs. In addition to saving sick or distressed crops, this information also enables farmers to avoid over-application of chemicals as well as over-irrigation, thus saving money and reducing environmental impact.
In every piece of technology in this futuristic scenario there are many components that require digital signal processing. All the different monitoring functions require sensors that can collect the relevant data, as well as communication processors that can send the data using various wireless communication standards to the cloud. From extremely low-power sensor networks, through proprietary communication for drones, all the way to high-speed LTE networks for tractors and concentrators, communication processors for Ag3.0 must remain flexible. The autonomous farming vehicles, tractors, combines, choppers and drones, all need vision capabilities as well as intensive processing power to derive intelligence from the visual data. They also require extremely accurate GPS capability, so that they can both navigate properly and report information on each and every plant that might need fine tuning in water, nutrient or pesticide application. The farmer, who is in charge of this whole operation, requires a handheld device with LTE and Wi-Fi connectivity and processing capabilities, in order to display the processed information, and make decisions in cases that require intervention.
Critical Challenges for Future Agriculture
The future of agriculture holds many critical challenges. Apparently, a big part of the solution is information technology and particularly the collection and analysis of data, data and more data. The collection and processing of these huge amounts of data require multitudes of efficient sensors, computer vision, communication, navigation and processing components, as well as ever-advancing software to translate all the data into useful information. Just as the farming industry is coming to the realization that precision is key, all the more so for the silicon industry. Every component in this procedure needs to be precise, “not a molecule wasted”, as Dr. Budzynzki stated. The embedded solutions that go into the tractors and drones need to be low on die size and power consumption and excel at processing and speed, so that the farmers can get the best and the most from their fields.
While Ag3.0 shows a lot of promise, it is clear that a significant capital investment is required to adopt these cutting-edge solutions. That means millions of smaller farmers are going to find themselves fighting against even larger conglomerates with deep enough pockets to continue the efficiency gradient. So with Ag3.0 we will see better efficiency, profitability, more produce with better quality, but also, perhaps, the elimination of the smaller farmers. How is this going to play out? Time will tell.
This post was also published on Embedded.com Click here to read
You might also like
More from Deep Learning
On Demand Webinar: Challenges of Vision Based Autonomous Driving & Facilitation of An Embedded Neural Network Platform
The automotive market is seeing accelerated growth and rapid adoption of vision applications that will lead the way to autonomous …
In two of my previous posts, I discussed deep learning frameworks, features and challenges and the specific challenges of deep …