IoT4Food: Harnessing Internet of Things Technology (IoT) for sustainable agriculture

The inability to obtain soil characteristics rapidly and inexpensively remains one of the biggest challenges faced by farmers in Uganda and many African countries. Currently, the majority of farmers do not use any scientific means to assess soil suitability for different farming activities. Farmers classify their soils based on indigenous knowledge and experience gained over several generations (e.g., soil colour and texture). However, these methods do not always give accurate information. The ambition of our research is to increase farm yield maximization for smallholder farmers through a cloud-based wireless sensor network of Internet of Things (IoT) for instrumenting soil metrics in order to provide real-time monitoring and data analysis for smart decision-making needed on a day-to-day and season-to-season basis. Our innovation encompasses development and deployment of a low-cost soil conditions monitoring technology that can be deployed up to the level of fields or farms by smallholder farmers. The technology informs farmers’ decisions regarding the best type of crop to plant and when to plant, when to apply fertilisers, when to irrigate and in what amount, among others. The technology is composed of two components: the hardware soil sensing devices and the cloud platform for soil analytics. The soil sensing devices are capable of providing data on soil attributes including soil moisture, pH, and temperature. These devices are deployed in gardens forming an Internet of Things (IoT). The sensors transmit the soil conditions data to cloud-based central system where data analysis, fusion and interpretation take place. Each device includes a Subscriber identity module (SIM) card that enables sending the collected data over any available mobile network. The cloud-based software platform is enriched with big data analytics for soil conditions to facilitate smart decision-making. The cloud platform accepts raw soil condition readings from sensors deployed in the farm. The collected data forms a big data repository upon which predictive analytics tools are deployed to analyze, interpret and process the data into high-level meaningful information for the farmers. The cloud platform is also capable of aggregating sensor readings from multiple data sources and thereafter providing users with personalized information e.g., an irrigation schedule with details of the specific amounts of water. Users can query the status of a given field (e.g., by test message using short message service, SMS to a specified number or using a mobile phone mobile application). The current devices and software platforms are undergoing further refinements so as to improve the accuracy of readings, packaging of the sensor, improving the battery life and energy consumption, and inclusion of other soil attributes as we prepare for scaling up the deployment.