Exploiting Novel Imaging to Optimise Crop Yield in Controlled Environment Agriculture
Supervisors:
Project summary:
Intensive agriculture is essential to feed increasing populations, yet requires large amounts of pesticide, fertiliser, and water to maintain productivity. One solution to mitigate these issues is the adoption of Controlled Environment Agriculture (CEA). The self-contained operation of these facilities offers the potential to recycle agricultural inputs, as well as sheltering crops from the effects of climate change.
Growing crops in CEA systems is now commercially viable, although the associated energy demands limit the choice of crops. One way to minimise the energy footprint of CEA is to vary lighting in parallel with the availability of renewable energy. In order to maximise plant growth with varied lighting we need to understand how plants adapt to these changing conditions. This project will use a combination of novel imaging methods to remotely assess plant performance. Machine learning algorithms will be exploited to predict final yield. These advances will enable farmers to adjust lighting regimes to maximise yield and energy efficiency in CEA systems.