Machine Learning and Drone Systems for Smarter, Welfare-Aware Agriculture
School of Engineering and Technology
Dr Ayub Bokani
Synopsis
Agriculture is being reshaped by artificial intelligence, drones and low-cost sensors, which together can watch over vast properties, track animals and crops, and turn raw data into practical decisions. This project applies machine learning, computer vision and optimisation to problems in smart, welfare-aware agriculture and autonomous aerial systems, from monitoring animals and pastures to planning efficient drone operations. The scope is broad and can be shaped to a student's interests and background, whether that leans toward data and modelling, computer vision, or the networking and energy challenges of keeping drones useful in the field. Current work in the group, published in journals such as Drones and at IEEE INFOCOM workshops, offers ready examples and real datasets to build on, including livestock movement prediction and cattle behaviour recognition from wearable sensors.
artificial intelligence, machine learning, precision agriculture, UAV, drones, computer vision, livestock monitoring, remote sensing
Available now (flexible)
Masters; Either Masters or Doctorate; Doctorate
Other Special Notes
Possible directions include: predicting livestock movement with deep learning, for example LSTM models, to plan drone flights and save battery; recognising cattle behaviour such as grazing, rumination and resting from collar sensors, and making models work across different animals; reinforcement learning for energy-efficient drone flight paths and wireless recharging; computer vision and remote sensing from aerial imagery for animal detection, pasture assessment and crop-health monitoring; and broader smart-farming problems in animal welfare, sustainability and precision agriculture. Current livestock-monitoring projects provide examples and datasets to start from.
