Submitted by admin on Wed, 08/10/2022 - 14:36

Non-uniform growth of crops impacting product quality and volume

Sequence Number 7 Industry Agriculture Banner Non-uniform growth of crops impacting product quality and volume How 5G enabled

Obtain crop growth rates through collection of data, monitor and identify areas where optimum growth is impacted, apply AI to identify possible reasons, and suggest actions to address problems.

Data Flows
Title Devices Icon Devices Description
  • Drones could monitor progress of crop growth
  • Sensors (fixed or on drones) to collect information, e.g. soil moisture levels, nutrients, pest, or any specific crop sensor
Title Connectivity Icon Connectivity Description
  • All sensors will transmit data to Enterprise storage
Title Edge Compute Icon Edge Compute Description
  • Not time critical at present, however this is expected to change to real-time in the future
Title Cloud Compute & Storage Icon Cloud Compute & Storage Description
  • All edge data collected is loaded into Enterprise storage
  • All data is stored at Enterprise storage
Title Applications & Services Icon Applications & Services Description
  • AI (MV) models provide information of current growth rates vs expected levels (predicted)
  • (AI – ML) Provide potential reasons if rate is below expected growth levels, and suggest actions
Title Inform Decision Makers Icon Inform Decision Makers Description
  • Reports providing visibility of crops, as model can operate as a closed loop
Title Support Decision Making Icon Support Decision Making Description
  • End of process
Application Logic
Description
  • Drones would be a key enabler to capture data over large areas to detect abnormalities, e.g. growth, pests, soil conditions, etc.
  • Begin data collection offline, however prepare for evolution to real-time once 5G becomes available when developing MV models.
Description
  • Development of AI (MV) models could include various other information (e.g., expected crop growth rate, plantation area, etc.) and produce results such as predicting growth levels for the foreseeable future.
  • Development of the MV model is done through an iterative process where multiple models will be required to detect and predict growth.
Description
  • The model created could start with simple predictions such as rain vs irrigation needs, detection of colour changes to crops indicating pests and could increase in complexity over time.
  • Joint effort between SME and Data Scientists is required for model development.
Expected benefits Key value created