Information and Communication Technologies and Robotics for Sustainable Agriculture
Fusion of multi-source and multi-sensor information on soil and crop for optimised crop production system

Project No: 14303

Project dates:
1 Mar 2013 - 29 Feb 2016

Abdul Mouazen, Faculty of bioscience engineering, Ghent University (Belgium)

Collaborating Institutions:

Dimitrios Moshou, Aristotle University of Thessaloniki (Greece)

Ralf Bill, Rostock University (Germany)

Yucel Tekin, Uludag University (Turkey)

Steffen Piecha, tec5 AG (Germany)

Jens Wiebensohn, Rostock University (Germany)

Project website
Full report


Proposal Summary

Ignoring the inherited spatial variation in soil properties with traditional sampling methods leads to poor crop management, yield loss and excess use of input. The proposed system of FarmFUSE addresses these issues in 2 ways: (a) utilising a new and innovative on-line multi-sensor platform for measuring key soil properties at high sampling resolution. (b) Integrating this improved soil data with other information such as vehicle-borne sensing of crop growth, weather data, soil conductivity and yield maps, to develop algorithms to determine rules for variable rate application. These information are then integrated into a FMIS. The final integrated system is a server that allows end users to access and upload data including maps of soil, crop, yield and variable rate recommendations.

Main Results

Results showed that the on-line multi-sensor platform to provide accurate measurement of key soil properties. The largest accuracy was reported for moisture content, organic carbon and total nitrogen. Moderate accuracy was reported for phosphorous, calcium, sodium, pH and cation exchange capacity. A new hybrid self-organizing map (SOM) and k-means algorithm for delineation of management zones maps showed better separation of clusters when compared with the standard k-means algorithm. The architecture of a FMIS was defined and a prototype is running. The cost-benefit analysis revealed FarmFUSE concept to be profitable to the partner farmers across the three experimental farms in Europe. Farmers’ feedback was positive about the project findings and supportive to become early adopters.


A FarmingTruth Ltd was established in the UK to provide services to the farmers at annually paid fees. However, this still needs further development to go through a demonstration phase before it becomes operational. A first generation of a tec5 partner’s embedded spectrometer platform was developed, and designated as tecSaaS platform, which was further developed under the CasiNIR R&D project. The final product is now commercially available since the end of 2016, and can be utilized not only in agriculture e.g., measurement of manure and soil but in other industrial applications outside FarmFUSE remit e.g., in the galvanic industry. In Turkey, Kayseri Sugar Factory, a big cooperative having about 10000 farmers is interested in establishing a precision agriculture service provider in Turkey.