Bee Monitoring and Behaviour Prediction
Collecting data on environmental factors and processes surrounding the bee-hive through IoT sensors, and training AI with these observations to improve predictability in beekeeping.
Successful beekeeping depends on multiple environmental factors and processes inside and outside the hive. To make beekeeping more predictable this Flagship Innovation Experiment (FIE) focuses on the long-term collection and monitoring of data from the surrounding environment and the hive itself such as temperature, humidity, sensor measurements, and expert surveys alongside observation data. Ultimately, Artificial Intelligence (AI) can be deployed and trained with the use of these insights. Hence, the main goal of this FIE is to deliver a technological platform, centralising all datasets in one entity. The monitoring is carried out by wireless sensor networks and similar Internet of Things (IoT) solutions.
As a result, notification systems for beekeepers can be implemented while irregularities are identified and resolved in real-time actions. To accurately identify changes in bees’ behaviours, the redesign and modification of existing implementations are a crucial part of this FIE. Thus, developers, service providers and end-users work closely together in this project to achieve the best possible outcome.
There are several lessons learnt during the implementation of this project. Dependency from 3rd party developed hardware solutions impacts planned development plans - readiness of used technology for use must be tested already in advance, reviewed as minimum one other completely independent alternative option. From today's experience and perspective, FIE should have one partner as a hardware provider with experience and flexibility to modify, adapt and deliver sensor nodes ready to use to FIE customers. Also, platform sustainability it's dependent on ready to use, simple and durable hardware solutions.