NPR - SmartFarming

Actual situation

The global food production system suffers from serious problems and contributes significantly to climate problems. In the future, despite farmers efforts to mitigate their impact, these problems will be further exacerbated by the significant increase in the world's population and changes in lifestyles. To illustrate this, it is worth mentioning that 70% of the world's fresh water is used by agriculture. Of this 70%, 60% of the water used for irrigation is wasted due to leakage and evapotranspiration, while many regions already face supply constraints.

Emerging of new solutions

Solutions for optimizing the use of resources and improved decision making in the agriculture industry are emerging today. However, many of these solutions are still at an early stage and return on investments are still difficult to establish. Also, these solutions are often too complex to be used by farmers efficiently and are generally much fragmented, targeting extremely specific problems. This lack of integration impairs data integration from different sources as well as in-depth analyses.

At the same time, the pressure from society and politics for developing more environmentally friendly agriculture is getting stronger and stronger. There is thus a need for developing solutions which are economically viable and well adopted by society and the professionals. This project aims at developing such a solution.

The goals of this project are:

-  On one side, it aims at gathering multi-modal data from different sensor types, ranging from standard soil sensors to imagery data. A sensor network using a long-range network infrastructure will be built on a production site and data will be collected on a private cloud infrastructure.

-  On another side, the data collected will be used for controlling and automating the rationale use of resources such as water or phytosanitary products. The system will allow producers to receive alerts based on specific criteria and to control or automate tasks such as water irrigation.

-  Finally, the project aims at studying existing and developing novel algorithms using multi-modal data such as soil temperature, moisture but also imagery. These algorithms will allow developing models for optimized use of resources, for a better prediction of plant diseases and easier decision making for the farmers.

The partners of the project bring together the entire production ecosystem and will contribute to the development of a solution that considers the multiple aspects of the problem.