Climatic control – with a curly tail

A computer that understands pig-language and can regulate climatic conditions in accordance with the pigs’ instructions. This is the ambitious goal for a joint project between the Danish company Skov A/S and CISS.

“Just imagine if we could create a system capable of monitoring and measuring a range of factors and then, on the basis of this data, we were able to ensure optimal conditions for the pigs.”

PhD Student Jens Alsted Hansen draws a pig on the board - with a big, handsome and curly tail. Then he starts to tell about the idea which he and a PhD colleague have been working on for about a year.

Animals as sensors

These days, temperature, humidity and a whole range of other climatic factors are measured in the stables in order to obtain the optimal conditions for the pigs. “We would like to expand this system to include observations of the behaviour and condition of the animals so that, in this way, the system will be capable of adjusting itself and ensuring that conditions are optimal,” he explains.

This project stems from the fact that stocks of animals at Danish farms have become so large that it is difficult and time-consuming for the farmer to keep an eye on all the animals.

“If a pig lies down it means, as a rule, that it is too hot, but if this pig is the only one out of a thousand pigs who lies down, it can mean that it is ill. This is what we need to be able to observe, and if there is a problem which the system cannot itself deal with, then, the farmer must be called,” explains Jens Alsted Hansen.

Benefiting the animals

It is SKOV A/S from Glyngøre, Denmark which is jointly financing the two PhD students’ work. A somewhat long-term investment perhaps, but one which offers considerable prospects.

“For a long time we have been convinced that there are technologies and theories in existence which neither we nor the industry are making use of and just as we finally decided to throw ourselves into researching a new technology, CISS emerged as an industry-related research centre. The basic concept is to integrate “good animal husbandry” into the automated system. In addition to bigger and bigger animal stocks there is, abroad, a tendency to use labour which is less and less skilled. Such a system will therefore not only be of benefit to the animals but also to the farmer, as the animals will develop more satisfactorily which, in turn, will lead to an increase in the quality of the meat,” says Martin Riisgaard-Jensen who is the Software Group Manager at Skov A/S.

From concept to reality

The partnership with CISS has been going on for a year and Martin Riisgaard-Jensen is extremely pleased with the results.

“We have had a good dialogue and we have succeeded in making things more concrete. Now we know where we are going, so the next step is to turn it into reality,” says Martin Riisgaard-Jensen.

The researchers have two more years in which to do this and at Skov, they are in no doubt that the project will be very beneficial.

“Of course, the project will not result in a finished product. At the very best we will have a prototype which can be matured into a product and at the moment I have no doubt that it will be successful. Irrespective of the results we will have gained a great deal of knowledge and experience with which we will be able to continue our work,” says Martin Riisgaard-Jensen.

Mathematical models 

It is exactly this transition from theory to practice which, as so often before, is the most difficult. The challenge is not so much based in gathering the diverse information. It is when all of these inputs, such as the behavioural patterns of the pigs, surface temperature, body temperature, sound and the absorption of feed, have to be calculated that it begins to resemble maths at a high level. The little pig on the board is quickly encircled by a myriad of complicated calculations.

“The pig has an effect on its surroundings in the same way that temperature etc. has an effect on the pig, and it is all these many connected factors for which we must create a mathematical model. We can begin with the fact that there is a wide range of so-called causal factors but, fortunately, we can eliminate a great deal of them. We know, for example, that the air temperature does not change much in the course of a few seconds, so all eventualities for which this would be a pre-requisite can be eliminated. We can, in the same way, eliminate a number of other situations which we know do not occur in the real world. In this way we end up with a number of possible situations with which we can work – though this still represents an enormous number,” explains Jens Alsted Hansen.

Big ambitions

Firstly the idea is that all information gathered from the different sensors will be continuously transmitted to a central computer which, in turn, will determine what needs to be done to ensure that the pigs have the optimal conditions. In the long run, however, our ambitions are bigger than this. “I can envisage each individual little sensor being a mini-computer itself that would be capable of participating in the total calculation. The small computers could, as it were, talk together and jointly determine what needs to be done. This would provide a much higher level of reliability and, in the long run, I am sure that such a system would be less expensive. In my opinion we will not be able to claim to have created an intelligent sensor network until we have achieved this,” says Jens Alsted Hansen.

FACT BOX   Skov A/S was founded by the brothers Kjeld and Kristen Skov at the manor house of the Østergaard Hovedgaard in Åsted, Denmark, in 1954. Since its foundation the company has grown to become one of the world’s leading producers and suppliers of components and systems for stable climate control. Today the company’s headquarters are in Glyngøre, Denmark, where there are 200 people employed and from where they serve a network of 75 distributors throughout the world.