Injecting manure autonomously
Robotic technology key to a better soil
The use of robot technology in agriculture has another significant plus, in addition to the usual advantages such as savings in labour, more capacity and lower costs. It has the potential to reduce soil pressure and thus improve quality. A consortium is active in Flevoland with that philosophy as its goal. The idea is to build an autonomous fertilizer robot that fertilizes smartly and conserves soil.
Dairy farmer Gerrit van den Pol from Dronten is faced with a challenge: he is looking for a manure tank for his company that has a high capacity but little ground pressure. The livestock farmer fertilizes a lot of newly sown grassland, where prevention of rutting and soil compaction is essential. Capacity and low weight do not go well together. This is how the idea of a smart fertilizer or fertilizer robot was born.
2018
deadline
Consortium autonomous manure tank
With his idea, Van den Pol approached Corné Kocks, lecturer in Precision Agriculture at Aeres Hogeschool Dronten. The idea of the autonomous slurry injector turned out to be a wonderful object of study. However, that requires knowledge of machine construction and robot technology. For this, machine builder Veenhuis and robotics specialist Precision Makers were approached. A consortium was set up with the cooperation of Veenhuis director Walter Veenhuis and Allard Martinet of Precision Makers. A second dairy farmer, Mark Havermans from North Brabant, has also joined. The goal: to have an autonomously operating and smart fertilization robot in the field by the end of 2018.
Start from practice
'The consortium is a melting pot of expertise', is how Kocks sums it up. 'You are best off with a practical situation. Teething problems can be tackled immediately. First, a package of requirements is drawn up, after which bottlenecks are identified that companies must solve individually.'
'Robotization gets its strength from repetitive operations', says Martinet. 'Think of welding, milking and feeding robots. The challenge is to apply this to manure injection. The ultimate wish is a machine that is careful with the soil. This is the result of a combination of low weight and the right tire combination. The next step is site-specific fertilization using NIR sensors and precision farming. A robot is not versatile. He can't work with 30 different machines like a tractor can. The first version may require another person to be present. He can then control several robots.'
Known components
What does the robot actually look like? On the drawing board, it concerns a tricycle with low-pressure tires, a tank of approximately 5 cubic meters and a slurry injector of 3 meters wide. Version 1.0 uses well-known components, such as a diesel engine and hydrostatic drive. Electric motors and solar energy are in the future.
Veenhuis: 'Applying slurry is a difficult job, because you have to transport large volumes. You use liquid fertilizer in much smaller quantities. The robot fills automatically at a filling station on the headland or in the yard. Transport is of great importance in this regard. As a manufacturer, we also look at commercial success. In our opinion, the robot has the potential to be used internationally. By having several robots work 24 hours a day, you can achieve a high capacity. It is interesting for contractors. They can focus entirely on manure transport. The robots drive it out.'
Path planning
In addition to the hardware, the software is also a huge job. That is a task for Martinet and Precision Makers: 'Applying manure smartly means filling smartly and determining your ideal route. We call that path planning. The machine drives in circles (loops) over the field and tries to cover as few meters as possible. An example: It may be more favorable not to fill the tank completely or, because of the field length, to fertilize only half the working width. You talk about artificial intelligence. For a human that is a piece of cake, but for robots enormously complex. As a driver you are constantly making decisions. Each in his own way.'
'A disadvantage is that the human driver also makes the wrong decisions', adds Veenhuis. 'Continuing in a wet spot, causing him to get stuck. By making decisions based on sensors, you can work more safely. The car industry is also working on that.' According to those involved, a drag hose system is therefore not possible. That method is even more complex. Van den Pol does not agree with this from the point of view of ground pressure, because a heavy combination is required for the pulling power. It goes up to 8 tons.
Safety not an obstacle
Another issue, which is high on the agenda of robot technology, is safety. This is easier to meet for agricultural applications than for self-driving cars. Martinet does not expect any issues from a legal or insurance point of view. 'The manure robot works on its own site, which makes it easier. We do have to make decisions about how much human intervention is required. Everything stands or falls with the path planning software. We use rtk-gps as a basis. That's reliable enough. The plot contours are loaded in advance, so that the boundaries are known. The issues are mainly related to the software. Building a prototype can be done relatively quickly. The preliminary phase is relatively long.'
Testing in 2018
However, Veenhuis sees a lot of work for the engineering from the manufacturing side. When the concept has been worked out and the path planning is ready, a prototype can be built. This will be tested in 2018, so that the results will be known by the end of next year and the machine will roll into its next phase. Consciously, only slurry is looked at. 'In the future, this may also work for solid manure,' Veenhuis thinks. 'We consciously don't make the project too broad, otherwise the development will take even longer. It is too early for a price tag, but we are keeping a close eye on the costs of the design and the costs per cubic meter of manure. However, benefits for the soil are difficult to express in monetary terms.'
Future vision
Lector Kocks already has a vision for the future: 'Version 1.0 is the standard machine according to a well-known concept: Diesel engine, hydraulics and slurry injector. Version 2.0 is energy neutral and in version 3.0 we use bio-based composite to build the machine. Solar energy or hydrogen may then be used as a power source.'
Niels van der Boom
© DCA Market Intelligence. This market information is subject to copyright. It is not permitted to reproduce, distribute, disseminate or make the content available to third parties for compensation, in any form, without the express written permission of DCA Market Intelligence.