The Food and Agricultural Organisation of the UN (FAO) predicts that the global population will reach 8 billion people by 2025 and 9.6 billion people by 2050. In order to keep pace, food production must increase by 70 percent by 2050.
However there are several barriers to fulfilling this imperative, including:
- The slow-down in productivity growth
- The limited availability of arable land
- Climate change
- The increasing need for fresh water
- The price and availability of energy, particularly from fossil fuels
- The impact of urbanisation on rural
labour supply – the average age of farmers is increasing with fewer young people going into the industry.
- In order to counter these challenges, the FAO recommends that all farming sectors should be equipped with innovative tools and techniques, particularly digital technologies.
Rather than replace farmer expertise and gut feeling, ICT-based decision support systems, backed up by real time data, provide information at a level of granularity not previously possible.
Precision agriculture is sometimes known as ‘smart farming’, an umbrella term for easier comparison with other M2M based implementations such as smart metering, smart cities and so on. Precision agriculture is a specialist methodology in itself. It is based on sensor technologies whose use is well established in other industries, e.g. Telematics for fleet management, environmental monitoring for pollutants, eHealth monitoring in patients and so on.
Precision agriculture aims to optimise the yield per unit of farming land by using the most modern means in a continuously sustainable way, to achieve best in terms of quality, quantity and financial return.Precision agriculture makes use of a range of technologies that include GPS services, sensors and big data to optimise crop yields. Rather than replace farmer expertise and gut feeling, ICT-based decision support systems, backed up by real time data, can additionally provide information concerning all aspects of farming at a level of granularity not previously possible. This enables a more precise understanding as to what is happening at grass roots level, in turn allowing better decisions to be made, resulting in less waste and maximum efficiency in operations. The disciplines and skills now required for agriculture include robotics, computer-based imaging, GPS technology, science-based solutions, climate forecasting, technological solutions, environmental controls and more. Hence to make the best use of all these technologies, it is essential to train farmers and farm managers in their use.
For all M2M implementations, IT systems gather, collate, analyse the data and present it in such a way as to initiate an appropriate response to the information received. For farmers and growers, a wide variety of information regarding soil and crop behaviour, animal behaviour, machine status, storage tank status emanating from remote sites is presented for action by the farmer. This is analysed in conjunction with a broad range of data. Farm offices now collect vast quantities of information from crop yields, soil-mapping, fertiliser applications, weather data, machinery, and animal health; these are all factors that influence farming such as soils, nutrition and weather. Data is the fundamental building block of smart farming, whether the data comes from a soil sample or a satellite correction signal. For example, data points collected can highlight both spatial and temporal variability within a field. Many factors can contribute to this variability; understanding the effect each factor has can only be measured and managed using statistical analysis of the data.
The chart to the right shows the different types of technologies involved in smart farming. The set of technologies used in smart farming is complex, to reflect the complexity of activities run by farmers, growers, and other sector stakeholders.
Finally, it is important to highlight the increasing important contribution of robotics in the smart farm vision through various types of autonomous vehicles, drones, automatic milking systems and fruit and vegetable picking robots.
Towards the Smart Farm
What makes precision agriculture special is the IT system at the other end of the supply chain, the decision support system at the back office. Whilst the technology is still in its infancy, the notion of ‘the connected farm’ is coming closer, particularly if the seven types of farming activity we have listed below are somehow connected not only to each other, but also to a raft of historical data such as weather events, climate, economics, product information and specifications, machine settings etc.
We also anticipate that the use of sensors in farming will spread to adjunct areas, such as environmental monitoring land management, and food traceability.
This is what the Internet of Things is all about, connecting systems so as to allow an integrated, multidimensional view of farming activities, enabling deeper understanding on how the whole ecosystem works. Precision farming would become ‘decision farming’.
It is also important to learn the lessons from other large scale ‘smart’ project rollouts, notably the smart metering projects ongoing in European countries.
From an industrial M2M perspective, the agricultural sector is still considered a minor sector. Whilst he more immediate impact of M2M technologies in agriculture is around providing remote connectivity between sensors in the field and farm information management systems, M2M technologies and all the technologies around the Internet of Things vision are key enablers for the transformation of the agricultural sector towards the smart farming vision. We also anticipate that the use of sensors in farming will spread to adjunct areas, such as environmental monitoring (see case study below), land management, and food traceability. This is a consequence of the greater public focus on issues such as food safety and wildlife preservation. It is also important to learn the lessons from other large scale ‘smart’ project rollouts, notably the smart metering projects ongoing in European countries. The UK government for one is taking great pains to ensure that a full regulatory frame- work exists to support the programme and that the full legal implications are understood. These touch on customer privacy, ownership of the data collected, and whether it is permissible for this data to be repurposed for other uses. These issues are equally relevant to the agriculture industry, and a similar framework needs to be implemented to reap the best advantages from ‘smart farming’.
In addition, there is the issue of the ownership of the data that is collected from the sensors: farmers are said to be reluctant to share their data. A business case needs to be made, together with some incentives to work with suppliers of feed for example and develop trust. Several EU initiatives in precision livestock farming, funded under FP7 and Horizon 2020, are looking in all those issues.
Application Areas and the Case of Precision Livestock Farming
The research has identified seven key application areas in which the Internet of Things vision can shape the farm of the future.
- Fleet management (tracking of farm vehicles)
- Arable farming, large and small field farming
- Precision livestock farming
- Indoor farming (greenhouses and stables)
- Storage monitoring – water tanks, fuel tanks
These application areas can coexist in several cases and they can also involve the use of different sets of technologies. For the sake of brevity, this paper will analyse the case of precision livestock farming as example.
Case study – Precision Livestock Farming
Livestock management is a ripe area in smart farming. There is a growing case for using the Internet of Things to monitor the health and state of farm animals, gather data and analyse that data and exchange data and analysis for a variety of reasons, there being different reasons for cattle, sheep, horses, pigs and chickens.
Milk has a high value, and robotic milking is on the increase. Today robotic milkers are all data linked by default. With software controlled milking in conjunction with the wearing of an electronic tag, farmers can gain detailed information regarding the yield and quality of the milk, but also the health and feed requirements of each animal, and warning of impending illness.
It is becoming increasingly apparent that the health of farm animals cannot be divorced from the health of the environment, land management and eventually, human health.
In the UK, the case for tracking in sheep farming is becoming more urgent. Here sheep farming is not uniformly distributed: particular breeds occupy specific environments to which they are best adapted. Hill farmers are often poor and their flocks wander over very large areas and it is hard to see if they go lame. In recent months sheep rustling has become seriously endemic, with farmers losing entire flocks. In the winter sheep can get buried in snow and may die if not found. In particular, lamb welfare and survival are critical, especially nearing the time for the ewes to give birth. Remote sensing of even parts of the flock would enable a better understanding of problems and suggest solutions.
Pigs and Chickens
Pigs and chickens stay mainly in their houses. Using monitoring techniques it is possible to ascertain housing conditions, not only controlling temperature and humidity but also measuring levels of dust and gases. It is also possible to monitor aggression and various behavioural parameters. Hence detailed information can reveal the state of the animal’s health and welfare, which in turn impacts the food industry further down the line.
In 2015, plans were mooted for the creation of a Livestock Centre of Excellence in the UK, backed by an initial £34 million investment. Supporters included multinationals and companies in the animal breeding, animal health, feed and nutrition and technology sectors, covering a broad range of the supply chain. The EU’s EIP-Agri Focus Group is also investigating emissions from livestock and their wider effects.