AFarCloud brings Smart Farming to everyone’s reach

In 2018, Spectral Engines joined AFarCloud, an international project cluster that aims to provide a distributed platform for autonomous farming. The platform will eventually increase efficiency, enhance productivity and animal health, ensure food quality, and reduce farm labor costs.


At the core of AFarCloud is a European consortium comprised of large multidisciplinary companies, machinery manufacturers and their subcontractors but also innovative small businesses that are looking for growth with their new product ideas. Research facilities all over Europe work as a combining factor for the whole consortium and are in charge of the non-commercial side of the research.


Mikko Himanka works as a project manager at the research and development unit of Centria University of Applied Sciences. Centria is a small community that has a significant impact on Finland’s applied research field. During the last few years, it has stepped to the forefront of European research funding receivers. As research mainstream often revolves around urban themes, the core players of AFarCloud, who had been working on previous projects together, concluded that a new agro-technological project was desperately needed. Centria was involved in the project from the get-go.


“In a nutshell, the main goal of AFarCloud is to find new ways of collecting, controlling and using data in modern agriculture. Traditionally, farms have functioned on inter-generational silent knowledge that has been passed on throughout the years. As farms have grown and commercialized, it has become crucial to be able to scale this knowledge to the needs of bigger farms and fields”, Himanka notes. “This will not happen without proper automated intelligent solutions. If there are hundreds of animals, it’s quite clear that they can’t be tended to individually. At the same time, production needs to be enhanced and streamlined. Sensor technologies, open digital platforms, and intelligent data refining are key factors in achieving this.”


As the end goal of AFarCloud will be a comprehensive cyber-physical system, cybersecurity is also an extremely crucial part of the project; digital security threats might pose a big risk and have physical consequences on animal welfare. “When we talk about all things cyber, we talk about the digital and physical world entwining. A modern farm with milking robots and automated working equipment is a good example of this kind of a cyber-physical system.”


Chain reactions

The companies and organizations working in the project all have a specific problem they’re trying to tackle. One of the partners develops new functionalities for drones that will create operational links with tracking collars on grazing livestock. The drone can then be used to check the health of individual animals. Similarly, sensor data that is produced by one partner will be further processed into valuable new information by using smart operations from another partner.


Spectral Engines’ role in the project has been to develop a specific solution for measuring crucial information when creating a formula for compound feed. “The analyzer itself works flawlessly. Achieving a more accurate, product-level calibration will require further development. However, the feed analyzer has already proved its worth, which is naturally crucial to the success of the entire project”, Himanka says.

Himanka sees many market opportunities for Spectral Engines’ solution outside the project as well. “Spectral Engines has been able to productize a sensor solution that will have a huge market demand in the future. Their sensors have sparked a lot of interest, and many organizations are eagerly waiting to utilize them. In addition to this, people working on the farms are interested in their feed analyzer and are ready to experiment with it, as it eliminates a major factor of uncertainty in the mixing of compound feed and silage.”


The first demonstration week

The achievements of AFarCloud will be on display in three holistic demonstrations in different European regions. The first demonstration was in October 2019 at a dairy farm called Kotipelto in Ylivieska, Finland. During the demonstration week, Spectral Engines showed with Mtech how to measure the required values from a feed store and process them into an updated recipe for mixing the feed.


“When several tons of feed get prepared at once, even a slight change in the mixture can translate into imprecision worth hundreds of kilograms in individual ingredients. Feed optimization is about saving money as much as it’s about being kind to the environment and the climate. Animal well-being and good profit margins are tied together, and in modern agriculture, their simultaneous management is crucial”, Himanka emphasizes.

According to the researcher, one project might not turn the whole world around, but it will be an important milestone towards more agro-technological research. “The challenges of modern farming are crystal clear: in twenty years, we have to be able to produce twice as much food on a much smaller area. It has to be done sustainably, and everyone who has the expertise to do something about it should take part. It has been nice to see that a company like Spectral Engines has done just so.” 


Developing the future of farms, one cow at a time

One person interested in saving the world and making a better future for animals as well as humans is Marko Sorvisto, a farmer, milk producer, certified agronomist, and the owner of Kotipelto farm. He’s a coveted speaker in agricultural seminars for his enthusiasm, knowledge, and expertise in animal well-being and the development of innovative farming solutions.


“At Kotipelto, we are very passionate about cow nutrition, health, and ecological solutions. The AFarCloud project combines all of these in a way that has never been done before, so of course, we wanted to join. Earlier, we’ve worked as a test farm in smaller projects, so AFarCloud felt like a good continuum for this. When Mikko Himanka initially contacted us, we were thrilled to participate in such an important project. Even though it came as a little surprise at first that we’re the only pilot farm in Finland, we’ve been able to rise to the challenge and be a vital part of this great consortium.”


Kotipelto’s goal is to expand to 200 hectares of field and 170 cows to accompany their three milking robots. The Sorvisto’s grow grass, barley, oats, and corn and have two regular employees as well as seasonal workers. Their cows live in a loose housed barn, and the farm produces more than 1,5 million liters of milk annually, making Kotipelto an exceptionally good place for piloting new technologies.


Quality food requires quality feed

Finnish farms produce the best quality dairy products in the world. You cannot achieve this status with poor feeding methods. At Kotipelto, AFarCloud aims at developing a feed and silage scanner that will enhance the production and quality of feed.


”I call it the Holy Grail of measuring protein and dry matter. When feeding cows, it’s crucial to know that the consistency of their food is optimal. In the end, the only thing that matters is animal well-being in all aspects. Milk production stops if the cow’s living conditions and feed are not right. When you have an old cow and a newborn calf in the same barn, everyone can’t eat the same thing. Our oldest cow Vanessa is 13 years old and just reached a huge milestone in her career – she’s produced 150 000 liters of milk. We could never accomplish this with an unhappy, or an unhealthy cow”, Sorvisto explains.

Feed's consistency varies as it gets collected from different parts of the field at different times. Previously feed samples got sent to a laboratory and getting the results sometimes took weeks. During those weeks, the consistency changed, making the acquired data no more relevant. “The silage scanner we’re developing in AFarCloud will revolutionize our whole industry. We will get accurate, real-time data, and we will be able to optimize what our cows eat instantly. Milk production requires a good microbial balance in the cow’s stomach. It is a delicate thing, and this analyzer helps us to maintain that balance. It will also help us keep the cows happy and healthy.”


Breaking barriers

There are many misconceptions about milk production in Finland, one of which is its impact on the environment. Compared to milk’s high nutrient density, its gashouse emissions are quite low. The Kotipelto farm takes environmental issues seriously. As the farm grows its own feed, the cows consume all carbon the crops have absorbed from the air and soil. The cows’ manure gets used as fertilizer in the fields.  “We aim for a fully closed carbon cycle and carbon-neutral feeding. In this day and age, it would be wrong not to think about the best possible solutions for the environment. To us, every cow is important, and so is the world we raise them in.”


According to Sorvisto, it is not only adult cows and the environment that benefits from optimized feeding. As feed has a direct impact on the quality and quantity of milk, it also has a direct impact on newborn calves. “The most important day in a cow’s life is the day they are born. It will determine the rest of their life, and to ensure that it will be a good, long and healthy life, we need to be sure that the calf drinks only the highest quality milk. At the end of the day, everything comes back to nutrition. Spectral Engines’ scanner is a cost-effective and powerful tool for all kinds of farms, big or small, for optimizing that nutrition. I think that there’s nothing more important than providing technology and help for those who produce food for all of us. This scanner will help the daily lives of farmers significantly.”

In the future, all data acquired with the help of AFarCloud, such as attained savings, carbon emission, and feed quality control, moisture levels, animal health checks with collars, and soil quality measurements should eventually end up in the same database. All of the innovations will hopefully work together seamlessly, with one software and one computer.


“There is still a lot of work to do before we achieve this kind of comprehensive model, but we’re definitely on the right track. We’re already so happy with the silage scanner that it might also be interesting to see how these kinds of solutions work at ‘the other end’ of the cow. Analyzing manure could bring us important data about how well nutrients absorb”, Sorvisto concludes.