Switzerland is characterized by valley, hill, and mountain landscapes. About one-third of the land is arable, while two-thirds is grassland. Plot sizes and farms tend to be small. Precision agriculture in the classic sense of sensor-supported management of grid or zone-specific differences in tillage, planting, fertilizer application, and plant protection is not widespread.
Profitable use of technology is primarily achieved by a contractor rather than by individual farms. Consequently, contractors are important agents for the spread of new technologies. Even though in practice precision agriculture is only implemented in individual cases, it generates a great deal of interest in both farming and research circles.
Targeting Weeds By The Plant
In Switzerland, agriculture that protects nature to the greatest extent possible, with the reduced use of herbicides, plays an important role. Finding an effective technique for controlling weeds in grassland is difficult, especially in organic agriculture where chemical plant treatments are banned. In particular, Rumex obtusifolius, also called broad-leaved dock, is one of the most competitive and persistent weeds in agriculture. The plant is very robust, and its dispersion is difficult to control.
Currently used methods such as time-intensive manual removal of the plants, manual treatment of individual plants with herbicides, and non-specific broad surface spraying do not produce satisfactory results. A Swiss research partnership is currently developing an automatic recognition and plant-treatment system as an alternative treatment technique.
An infrared-laser triangulation sensor and a high-resolution smart camera are used to generate 3D images of the weeds and their natural environment. In a segmentation process, contiguous surface patches are separated from one other. These 3D surface patches are compared with different criteria of a plant database containing surface parameters such as shape, state of surface, etc. When objects are extracted and confirmed as possible leaves or parts of leaves, their texture is analyzed by comparing the information of simultaneously taken 2D images with database criteria. If an object is recognized as a dock leaf, its coordinates in the vehicle coordinate system are computed and the leaves are sprayed with herbicide. The surface analysis in space can boost segmentation performance under conditions where state-of-the-art 2D recognition systems are not successful, e.g. low contrast, green-on-green images, noisy images, or images taken from inappropriate positions.
Initial results have been promising. System development focuses on a more robust imaging-sensor technique and refinement of the different algorithms. Looking to the future, the system design allows for the flexible integration of other plant species.
Controlled Traffic Farming
The soil-friendly wheel traffic concept of Controlled Traffic Farming (CTF) using GPS steering is already used extensively in Australia in field and row crops as well as in vegetable production. GPS steering systems ensure that the same vehicle tracks are used time after time, and as large a proportion of the land area as possible is no longer subjected to wheel traffic.
Although GPS steering systems are spreading more and more throughout Europe as well, CTF methods must still be adapted to the basic local strutural and technical conditions. In a field trial geared to Swiss conditions, a modified wheel-traffic concept is being examined and further developed. The hypothesis is that in addition to improving direct-drilling methods, CTF can also bring benefits in grassland management and vegetable growing.
For economic reasons, contractors are compelled to utilize their machines as close to capacity as possible. GPS-supported planning and documentation solutions with continuous electronic data transfer between office, driver and machine help with job handling, billing and analysis of cost structures, as well as with the fulfillment of traceability obligations. Agricultural-machinery cooperatives and cooperative agreements between contractors with high social-welfare standards can be more easily set up by means of an IT system that fosters trust.
As proof of good professional practice and as a prerequisite for drawing direct payments, farmers must furnish extensive documentation for both plant production and the animal husbandry associated with it. Data is used not only in-farm but often also cross-farm, by the State Administration or buyers/processors.
In both cases, the creation of uniform communication standards such as ISOBUS for tractor-equipment combinations, ISOagriNET for in-farm technology linking, and agroXML for inter-farm data exchange with all agricultural interest groups is important. With communication standards as a basis, food-chain information systems can be built up. This information linkage holds great potential for the food economy as a whole.