INTEL-IRRIS - Intelligent Irrigation System for Low-cost Autonomous Water Control in Small-scale Agriculture

According to FAO, small-scale farming has an enormous contribution to food security and to rural economy. However, smallholders usually face a number of constraints that are impeding their productivity, profitability and contribution to economic growth. Water resource is one of the major constraints and the situation is foreseen to worsen due to water shortage in relation with current excessive use and climate change. Controlled and improved irrigation can save water while maximising plant growth and yield. The usage of smart technologies and especially sensor systems – for instance soil water content and matric potential sensors – is not new in agriculture. So-called “Smart Farming Technologies (SFT) cover a range of different aspects of precision agriculture including data acquisition technologies, data analysis and evaluation technologies and precision application technologies”. When it comes to irrigation in particular, the underlying approach is to collect real-time soil conditions and to schedule irrigation when needed, possibly taking into account various data sources such as soil texture, soil chemical properties and weather data to name a few. However, existing high-end commercial systems are mostly based on very high cost professional sensor devices (usually coupled with costly soil mapping process) which dramatically limits the number of units that can be deployed. On the other hand, when adopting a low-cost design approach, one main issue is the reliability and accuracy of the collected data which is always low and can dramatically limit the efficiency of the deployed system.

Intel-IrriS will propose a low-cost water control system but will improve its efficiency. It will do so (1) by enabling deployment of several complementary low-cost sensors – and not only one costly sensor for a large area – to take into account the soil spatial and water spatial & time variability at field scale, which is made possible by the much lower cost per unit, (2) by using automatic and remotely controlled procedures for advanced calibration of the different sensors to increase accuracy of collected measures so that our water-soil-plants-climate interaction models will provide increased accuracy on recommended actions and (3) by including agricultural models/knowledge with corrective & predictive analytics – from simple computer-based decision models to more advanced AI-based processing – to adapt the applied control to local conditions & practices (dry region, open field or greenhouse) and crop/plant varieties that usually have different water need profile at their various stage of development.