Farming for the future: how technology is doing good down on the farm

Most people would agree that the purpose of technology should be to assist its users to live a better, more comfortable, easier life. And to have positive impact on a personal, societal and global level. Solving the looming food crisis predicted by the UN Food and Agriculture Organisation (FAO) among others would be one way of achieving this goal. It is predicted that by 2050 the world population will have reached 9.1 billion people all needing food to survive. However, the FAO predicts that food production would need to increase by 70% to feed the population. Another compounding factor will be the economic rise of Asian countries with increasingly large middle-class populations demanding more resource intensive food products such as pork, beef and dairy. In addition, Innovate UK predicts that climate change will reduce crop yields by 5–10%. In what has been termed Ag 3.0 the farming sector is reacting to this pressing need by incorporating cutting edge technology from the fields of AI, robotics, IoT, autonomous vehicles and drones – a process that Goldman Sachs believes will increase crop yields by 70% by 2050. This article looks at some ways this is being implemented.

Entirely automating the farming process would significantly increase the speed and efficiency of agriculture, in the same way that the invention of agricultural machines such as the threshing machine or the combine harvester did – both reducing labour demands and increasing efficiency. In 2017 a project was started at Harper Adams University in the UK called the ‘hands-free hectare’. The project demonstrated the potential of autonomous farming by harvesting a field of barley. In Spring 2017 the team grew a field of barley using autonomous vehicles, the process from pre-drilling to harvest involved 10 different steps including:

  • spraying herbicide and crop sowing using an autonomous tractor
  • using a small robotic rover to monitor soil health and moisture
  • during the growing period, using a drone to take aerial images of the crops to assess health
  • an unmanned combine harvester passing over the field collecting in the crop.

The team now plans to improve its technology for future projects, including working on increasing the autonomy of the tractor and other vehicles so that they can drive from shed to field, as well as improvements to their vehicle-to-everything (V2X) systems to allow them to interact with other objects such as a field gate, enabling it to open and close on command. Despite the successes of this project, an entirely autonomous farm may still be quite far off. Below we look at some solutions taking place in different parts of the agriculture sector.

Robotics and autonomous vehicles

Cereals can cope with some rough handling making them suitable for the types of robotic harvesting outlined above. However, what about fruit which not only needs careful handling but also the expert eye of a human to assess its ripeness?

While remaining a challenge for roboticists there are a number of companies working on systems which pick soft fruits and vegetables such as strawberries and tomatoes. Based in Spain, Agrobot is one of those companies, and in addition to assessing ripeness and picking, can even remove the stem or calyx from a strawberry. Its E-Series harvester uses LiDAR-based navigation to move along the strawberry rows picking off ripe fruit with its 24 robotic arms; each fruit is then deposited in boxes arranged under the robotic hands. The harvester uses graphics processor units (GPUs) and 3D computer vision (based on infra-red depth sensors and short-range colour sensors) to assess whether individual strawberries are ready to pick.

Japanese company Panasonic is concerned with the demographic issues facing the East Asian nation. It is particularly concerned with the lack of farm labour this causes and the growing lack of sufficiency in the Japanese food market. It believes that robots could be the answer to some of the problems Japan, and the rest of the world, faces. It has invented a tomato picking robot – tomatoes being one of the most labour-intensive fruits to farm. Its bots have been trialled in intelligent greenhouses that monitor and adjust levels of temperature, humidity, light, water, fertiliser and carbon dioxide to optimise plant growth and yield. According to the trial farm’s manager, harvesting the tomatoes traditionally accounts for 20% of the entire work load: 35,000 hours. The robot is attached to a rail that follows the edge of the tomato plant rows. It moves along the rows and uses a camera with image recognition to see which tomatoes are ready for harvesting. Those that are deemed ready are gently lassoed and cut from the stem from where they fall gently into a bucket below. The current speed is one tomato every six seconds, which is slightly slower than humans, but it has the advantage of being able to work all day every day.

Another Japanese company, Yanmar, has recently released an autonomous tractor. The ‘Robot Tractor’ can autonomously perform functions such as turning at the end of a row in forward drive mode and also in reverse. The autonomous vehicle can be controlled through a 10-inch tablet allowing operators to remotely drive two tractors at the same time. There is also the option to retrofit the autonomous gear to an existing tractor. Yanmar states that it plans to sell 100 tractors per year.

The hope for autonomous vehicles on the farm is not only automating the process but also increasing yield through a fleet of lighter vehicles working fields, rather than one heavy, human-driven machine that causes greater soil compaction and reduces yield by 13%.

Big data and the IoT

Another area opened up by the large amount of computing power available today is analysis of data from sensors and other sources, and consequent actionable insights. The potential is being tested by UK organisation Agrimetrics, an agricultural data analysis company partnered by the UK Government, and academic and industrial organisations. One area of interest is meat production efficiency. With the rise of the middle class in countries such as China, India and Southeast Asia the demand for meat will create increased pressure to produce it in large quantities. Agrimetrics, with other AgriTech Centres, is developing a system to combine pig and cattle data from connected sensors to improve meat production. It will consider numerous factors such as temperature, humidity, wind and solar radiation and its effect on the quality and quantity of carcasses. This information can then be used by farmers to increase output.

In the arable and market garden sector, companies such as Cropio have developed holistic management software that presents collated and processed farm data on a dashboard, helping farmers see field history, current statuses, farm machinery real-time movement information, crop density, alerts, soil moisture and harvest forecasts, among other information.

IoT has also been used to track and monitor livestock helping to keep it healthy and reduce wastage. Examples include Huawei’s Connected Cow solution that monitors cow fertility periods to ensure optimum mating time to continue producing milk and alerts the farmer through an NB-IoT network when the cow is in oestrus. Telecom provider Telia connected flocks of sheep to an NB-IoT network during the summer of 2017. The monitor consisted of a transmitter that is worn on a collar round the animal’s neck to enable farmers to remotely watch over their flock’s movements and alert them to any aberrations that could indicate sickness or injury. Moocall, based in Ireland, sells a sensor that attaches to a pregnant cow’s tail and alerts the farmer when the cow goes into labour enabling him or her to assist with the birth and reduce potentially fatal complications. It has also recently released Moocall Heat, a collar-based tracker that monitors when cows are in heat so that farmers can take appropriate action.

Other uses of IoT involve monitoring food as it moves along the chain from farm to dinner plate to reduce food wastage – a figure that stands at 40% in the USA according to the National Resources Defense Council. Fridge and container temperature and humidity monitors can help keep food fresher for longer as its transported around the world, driven to retailers, or stored in warehouses or restaurants. The use of blockchain technology – an unalterable distributed ledger – also offers chances to reduce wastage and identify problems in the supply chain. It allows customers and businesses to check where their food came from and where it went on its journey to them. This is done through a network of data sources that are effectively hack-proof ensuring the data presented is the truth.

Another area where IoT and big data analysis can have tangible effects is in the area of water usage. While water is a relatively cheap resource for farmers – compared to say labour – the agricultural industry uses 70% of the world’s freshwater, an issue that will come to fore in the near future as fresh water shortages lead to more widespread water shortages and possible political disputes. The effects of water shortages and the increasing cost of irrigation are currently being felt by farmers. In California – a traditionally productive area – farmers are switching crops from water intensive citrus fruits and avocado to less water-hungry crops such as dragon fruit and pomegranates that have lower yield and sell at a higher price in shops. This will not help the food problems – only 140 years ago California was the second largest producer of wheat in the USA. There are a number of systems which can optimise watering schedules and increase efficiency at the same time. The two main solutions are weather-based and soil moisture sensors. The former uses weather data to determine watering schedule. The system uses temperature, wind, solar radiation and humidity. The latter system uses soil moisture sensors to trigger watering when it senses low moisture levels. According to Hydropoint, a smart irrigation provider, using these systems can lead to water savings of between 30 to 50 percent.

Drones

Drones have a wide application in covering large fields for monitoring and spraying – inadequate fertiliser application can reduce crop yield by 15-20%. Companies are selling drones purpose-made for farm work. DJI has a drone for spraying crops with fertilisers or pesticides: it is an 8-bladed copter drone with an attached 10 kg liquid tank payload. DJI claims it can autonomously spray an area of 4000-6000 m2 in 10 minutes, which is 40 to 60 times quicker than manual methods. It allows precise control with settings choices that allow the user to decide on spraying speed, height and drone speed.

Drones are also finding use as eyes in the sky helping farmers gain insight into crop growth and about microclimates within individual fields. Multispectral imaging enables farmers to control crops, soil, irrigation and fertilisation with increased efficiency. The premise is that different internal components of plants such as chlorophyll or beta-carotene reflect light differently enabling easy identification of specific issues that would affect those readings. Using data from the images captured by drones equipped with hyperspectral cameras and through specialised software farmers can see pest activity and plant disease and then act with an appropriate pesticide regime; estimate crop yield; identify areas lacking nutrition and act with appropriate amounts of fertiliser; survey fences; report on microclimates and monitor livestock.

Figure 1: Where technology is being used to increase food production

The advantages for farmers of using connected sensor networks to save money, labour and time will continue to provide impetus for them to continue adopting and integrating the technology into their farms. This will have a knock-on effect of helping ease the expected food shortages mentioned at the beginning of the article. Although agriculture may not be the sexiest application of technology it does demonstrate how technology is being used for good. When technology can be beneficial for users and mankind at large, this should receive admiration; headlines of horrifying what-if scenarios may attract more click-based traffic, but the good that technology can achieve should be commended and more widely circulated to the general public. It is only by doing this that we can avoid defensive, fear-based rejection of technology. Technology can help humanity facing global issues that could affect everyone and planetary life itself.

Add this: