From Precision Agriculture to Prescription Agriculture
According to UN reports, global population will increase by 2.5 billion over the next 30 years. At the same time, the number of people living in cities globally will increase from 50% today to almost 70%. Combined, it means that significantly more food will have to be produced with less people involved in actual food production. Not only is demand for greater quantities of food increasing, but production conditions are becoming more stringent, adding another layer of complexity to agriculture.
Another important aspect that has to be taken into account is the intensive use of natural resources in agriculture, water being an excellent example as 70% of global fresh water consumption is used in agriculture. Sustaining the current way of doing agriculture, in the light of significant increase of global food production, is not possible. Combating these challenges, doing more with fewer resources, requires a new approach to food production. This new approach has to ensure not only more efficient agriculture, higher yields and higher crop quality, but also careful use of natural resources and improved efficiency at all steps of the complete food production and consumption chain, including logistics, retail and finally recycling.
Agriculture is already today heavily reliant on technology; agricultural machinery is used throughout the process to automate or expedite different steps of agriculture; without it, modern agriculture would not exist. Over the last ten years a genuine uptake of precision agriculture has taken place in which data from satellites, sensors (including laser scanners, radar, thermal, multispectral and hyperspectral imaging), and global navigation satellite systems are used to reduce inputs, increase efficiency of the cultivation or husbandry process, and thus increase farm incomes (Gebbers & Adamchuk 2010).
These developments were to a large extent focused on enabling high-end machinery like tractors to gather data about the ongoing agriculture process (tracking location, speed and collecting information from sensors attached to the tractors and machinery) and optimize their operation (e.g. movement, harvesting, planting, fertilising) with minimal human supervision. Recent technological advances are simplifying design and deployment of such systems thus enabling wider adoption. LPWAN technologies are playing a very significant role as is enabling long-range connectivity for battery powered devices deployed across fields which is, together with robustness and low maintenance, one of the crucial requirements. Advances in sensor technology are making new in-field measurements possible, opening up opportunities for a range of new previously unfeasible services. IoT platforms are making it easy to collect, store, process and visualize data.
Of course, there are still many challenges to be solved. In the sensor design domain, in addition to enabling various in-situ measurements to replace costly and cumbersome sampling and laboratory measurements, ensuring reliable and accurate measurements in different environments (for example different types of soil) is one of the significant challenges. The ability to remotely (or automatically) calibrate sensors will be one of the crucial factors that will determine the impact and success of smart agriculture solutions.
The second and probably most significant challenge lies in data, in particular the science and engineering of data analysis and data manipulation. The ability to share data in a simple and efficient manner between different systems to enable extraction of relevant knowledge beyond what is possible in isolated systems is the most significant challenge. The market of precision agriculture solutions is currently highly fragmented, and has been created as a series of closed systems which do not enable simple reuse of data in other applications. Even when local systems are in place, data from one sensor platform (or a service) is often incompatible with data from other field resources, and there is no simple way to integrate them to extract valuable "correlated and actionable" real-time information. The combined challenge of analyzing these incompatible and now large, growing, and disparate data sets, as well as a fundamental inability to correlate data and create actionable events, has resulted in an agricultural technology environment that has failed to maximize the potential value of smart, connected products.
Associated with the previous challenge is the availability of digital agricultural enablers, i.e. readily available digital components that represent digitalized agriculture expertise in an easily used form (accessible as web services for example) that can be combined with other IoT services to create new smart agriculture applications. The availability of such enablers would allow more rapid and widespread development of smart agriculture solutions leading to the democratization of agriculture by making agriculture expertise available to everyone in an easy to use and consume format. It would also open new opportunities for agriculture experts to address global markets and create an easier mechanism for converting extensive theoretical knowledge into a global product. Machine learning and advanced data analytics algorithms will play an important role towards the creation of full digital farming enablers capable of processing inputs gathered by various systems and potentially leveraging learnings from global agriculture production.
The next step in the design of smart agriculture solutions has to be in the direction of open systems, reusable services and the creation of widely available agriculture platforms that can serve as the basis for development of new solutions, adapted to different regions, different crops, machinery used, etc. The creation of such interoperable systems will be driven by wide adoption of open interfaces with standardized data models, supported by common semantic data annotations.
Initial steps in this direction are being taken in different forms. The European Commission has funded a significant number of projects in the smart agriculture space through the FIWARE accelerator program and is continuing to do so through a large scale pilot program where one of the five focus areas is agriculture. Several startups are offering expandable solutions, often starting with farm management, but with the capability to add new services, even from third party vendors. Large multinationals are counting on data analytics and machine learning algorithms and are moving towards building new services or supporting enablers. Agricultural machinery vendors are developing their own ecosystems using in-house capabilities or through acquisitions. Global cloud providers, together with partners, are looking into adding agriculture specific features into the growing list of supported functions.
Slowly, agriculture is taking the main stage from smart cities and smart homes when it comes to the potential benefits users can reap from using IoT, cloud and data analytics based solutions with strong agricultural expertise embedded. Agriculture is moving from precision to prescriptive, helping farmers to know in advance what to expect and how to take the best measures at a given time to increase quality and yield, thus increasing profit while at the same time addressing the important environmental (reducing the usage of pesticides, water savings) and health aspects (more ecological food).
R. Gebbers, V. Adamchuk: Precision Agriculture and Food Security,Science 12 Feb 2010: Vol. 327, Issue 5967, pp. 828-831 DOI: 10.1126/science.1183899
agroNET agriculture suite: http://agronet.solutions/#
Accenture digital agriculture solutions: https://www.accenture.com/us-en/insight-accenture-digital-agriculture-solutions
H2020 IoT-01-2016 Call for Large Scale Pilots: https://ec.europa.eu/research/participants/portal/desktop/en/opportunities/h2020/topics/iot-01-2016.html
F. Guerrini: The Future of Agriculture? Smart Farming: http://www.forbes.com/sites/federicoguerrini/2015/02/18/the-future-of-agriculture-smart-farming/#48a88996337c
Dr Srđan Krčo is a co-founder and CEO of DunavNET, a company designing turnkey IOT solutions. He has over 20 years of experience, working with large multinational companies and international collaborative research and innovation projects. Currently, he is coordinating the research and innovation H2020 project TagItSmart, creating enablers for IoT of mass-market goods. Srdjan is one of the founding members of the International IoT Forum and is actively participating in IERC (IoT European Research Cluster), IoT-EPI (IoT European Platforms Initiative) and AIOTI (Alliance for IoT Innovation) activities. In 2007, Srdjan won the Innovation Engineer of the Year award in Ireland. He has published over 15 patents and more than 70 papers at international conferences and journals and is a frequent speaker at international events addressing IoT and its applications. Srdjan is a Senior IEEE member.
Dr Boris Pokrić, DunavNET CTO and co-founder, obtained his PhD degree in Artificial Vision Sciences in the UK. He has over 20 years of experience in the ICT sector, of which 12 years in the UK. Boris has managed off-shore and local teams in Openwave (UK) and Myriad Group (France) working on projects for large customers such as Motorola and Samsung. His interests are in the areas of Augmented Reality and IoT and he is engaged across multiple activities within the company such as business development, product management and system architecture.
Subscribe to the Newsletter
Join our free IoT Technical Community and receive our Newsletter.
Calendar of Events
2019 IEEE International Conference on Internet of Things and Intelligence System (IoTaIS)
5-7 November 2019
"IoT and the mmWave Frontier", an IEEE IoT Vertical and Topical Summit, co-located with IEEE RWW2020
26-27 January 2020
San Antonio, TX, USA
IEEE 6th World Forum on Internet of Things (WF-IoT)
5-9 April 2020
New Orleans, LA, USA
Call for Papers
Special Issue on Internet of Things for Smart Ocean
Submission Deadline: 1 November 2019
Special Issue on Software Engineering Research and Practices for the Internet of Things
Submission Deadline: 15 October 2019
Special Issue on Deep Reinforcement Learning for Emerging IoT Systems
Submission Deadline: 1 October 2019