october 2020 – september 2024
There has always been variation between fruit producing plants, with respect to the plant’s genetics, its environment, and the management of the environment. We believe that there is a link between this variation and the amount and quality of fruit produced. Therefore, we will be using AI and Deep Learning techniques to process environmental variables such as temperature, humidity, light intensity, and soil moisture. This processing will be focused on identifying the underlying patterns within the observed data, using expert knowledge of growing conditions to constrain the model’s intuitions.
The main objective of this project is to develop and deploy a novel AI system that can forecast a tunnel environment, and use that tunnel forecast to identify areas of the tunnel which may need greater observation for plant stresses. Thus, the main systems to be developed as part of this project can be summed up as follows:
1. A system that given several historic environmental factor observations, produces a heatmap that covers the tunnel with estimated values for that environmental factor, with respect to tunnel configuration
2. A system that can forecast to be within a tolerance of +-3 units with a 3-day lead.