Golf Course Maintenance Neural Network

The main purpose of the neural network I want to create is twofold:

  1. Nutrient Status Map: Soil core samples will be taken on a grid system every 5 meters. Those core samples will be sent to a lab where their nutrient content will be determined. The neural network will analyze the nutrient data and GPS location information of each soil core to generate a color-coded map that indicates the nutrient status of the soil across a golf course. This map will show a gradient from “not present” to “optimal” for each nutrient, tailored to the specific types of turfgrass present in those different areas.
  2. Fertilizer Application Schedule: The system will need to have two options: Either the user will input a list of commercial fertilizers which the already have on hand, or the system will recommend a fully optimized schedule in the case that the golf course managers are open to ordering fertilizers. I’m not sure if that means the neural network will need to be trained on every single nutrient label on every single fertilizer commercially available. The neural network will then provide an optimal soil amendment schedule. This schedule will recommend how and when to apply fertilizers to adjust the soil nutrient levels to their optimal values for the health and maintenance of the turfgrass.

To achieve these objectives, the neural network will need to be trained on data points that include the optimal soil nutrient levels for various types of turfgrass. It will then use this training to interpret real-time soil data and fertilizer information to provide actionable insights for maintaining the golf course turf. The neural network will likely be a form of a deep learning model, such as a convolutional neural network (CNN) or a preference neural network (PNN), as these have been used in related fields for tasks such as diagnosing nutrient deficiencies in plants. The model will require a dataset that includes the optimal levels of nitrogen, iron, oxygen, phosphorous, potassium, boron, copper, manganese, molybdenum, zinc, chlorine, magnesium, sulfur, calcium, and nickel for each type of turfgrass.
Given the complexity of the task, the neural network will need to be robust and capable of handling multi-dimensional inputs, as well as producing detailed outputs that can be easily interpreted by golf course managers for practical application. I myself am a golf course maintenance manager, so I will be able to help with that part. Does this interest any of the experts here as a collaboration project?
I have the training data set.