Submitted by admin on Wed, 08/10/2022 - 14:12

Non-optimal pesticide usage

Sequence Number 5 Industry Agriculture Banner Non-optimal pesticide usage How 5G enabled

Collect pesticide usage across the plantation and manage usage based on actual crop needs.

Data Flows
Title Devices Icon Devices Description
  • Sensors/drones to collect time series data on crop status to discover pesticide needs
  • Sensor for other data: growth levels, soil moisture, etc.
Title Connectivity Icon Connectivity Description
  • Crop (timeseries) sensor data transmitted
  • Other relevant data
Title Edge Compute Icon Edge Compute Description
  • Not time critical at present, however this is expected to change with availability of equipment with real-time pesticide control
Title Cloud Compute & Storage Icon Cloud Compute & Storage Description
  • All edge data collected stored to train AI algorithm
  • All data stored in Enterprise storage
Title Applications & Services Icon Applications & Services Description
  • AI (ML/MV) models used to:
  • (MV) detect pests with drones
  • Trigger the right level of pesticides to be applied to by sprayer machines in identified areas
Title Inform Decision Makers Icon Inform Decision Makers Description
  • End-to-end visibility on crop status, and pesticide needs/usage for real-time preventive action
Title Support Decision Making Icon Support Decision Making Description
  • End of process
Application Logic
Description
  • Sensors and drones are used to monitor crop statuses: 
  • Tracking growth levels
  • Crop changes, identifying pesticide needs
  • The use of agriculture sprayers for pesticide application can start offline but will move to real-time (5G) where the amount of pesticides sprayed can be controlled in different areas.
Description
  • 5G becomes relevant when we want to create a real time loop: Drones collect data about the field, pass the data to the AI (MV) application to scan the results and these will be used to control the provision of pesticides.
  • Without 5G, the required bandwidth will not be sufficient to get the data in real-time to the AI (MV) apps for verification.
Description
  • AI (MV) models about predicting pest levels (and crop growth levels) and subsequent pesticide levels (and type) to be applied.
  • Build up the MV model in steps using multiple SMEs and Data Scientists to get to end-to-end MV/ML models.
  • Over time, look at what other options are possible to further reduce the need for pesticides.
Expected benefits Key value created