Research platform to study sheep behavior

Abstract

We present the design, manufacturing, and testing of an online sheep behavior monitoring research platform for extensive livestock conditions. This kind of system can contribute to animal well-being and assist farmers in decision-making for better productivity. Our proposal comprises a wearable electronic collar device and a cloud server that stores data and provides a user interface. The device has an Icarus IoT Board from Actinius, which allows for motion data collection with a three-axis accelerometer, location data measurement, Narrowband IoT communication, and two portable power sources: solar panels or battery, both in the collar. Our application acquires encoded accelerometer data at 100 Hz, location data obtained every 10–30 seconds, and battery and cellular signal level data every 50 seconds and sends them to the cloud server. We use the Amazon Web Services as a cloud server for deploying our software solution. We manufactured thirty collars which are currently under test. The devices successfully collect data and send it to the cloud server. Cloud data can be remotely accessed and viewed via a web-based user interface. The research platform allows the implementation of data processing techniques, like machine learning algorithms, to perform behavior classification, both on the collar and on the server side.