- Project Title: Analytics for animal health service provision
- Partners: Centre for Intelligent Dynamic Communications (CIDCOM) University of Strathclyde and Afimilk Ltd
I am an EngD candidate using machine learning to analyse cow behaviour and health for optimised farming. This is enabled by the University of Strathclyde, where I previously earned my Masters in Electronic and Electrical Engineering, and industrial partner Afimilk Ltd, who have a strong history in developing support systems for farming.
My main interest is acquiring knowledge from complex systems and using machine learning to give computers the ability to act on this knowledge. This gives many opportunities for exploitation and interesting discovery, such as personalised healthcare, language processing, self-driving cars, game AI and music analysis.
Smart farming and automation is an ever increasing aspect of modern commercial farming. We aim to provide decision support systems for dairy and beef cattle which integrate a range of sensor data to provide real time updates to the farm staff both on system on farm and the cloud. This enables farm staff to get real-time analysis, health alerts and potentially recommendations to improve animal health and performance across the farm.
Using the Silent Herdsman system, provided by Afimilk, we can get detailed insight into the rumination and eating patterns of the animals. By modeling the expected behaviour of these animals, we can detect deviations from this baseline and provide oestrus alerts, enabling the farmer to optimise his herd’s fertility. This is being expanded to include detection of lameness, mastitis, and other key health concerns of cattle farming, as well as looking at the growth characteristics of beef cattle to promote an optimal growth strategy.