Summary
An agricultural operator managed large, distributed fields and wanted better insight into soil moisture and temperature, local weather conditions, and crop status in hard-to-reach areas.
WiseVision's stack enabled:
- LoRaWAN sensors integrated into WiseOS
- Drone flights coordinated via MCP ROS2 to verify anomalies
- A simple Digital Twin of fields and zones for planning and analysis
Stack Used
- WiseOS
- MCP ROS2
- Digital Twins
- LoRaWAN
- ROS 2
Key Benefits
- Better targeted drone inspections
- Understanding of field variability over time
- Scalable pattern for new fields
- Visual spatial context
Challenges
Wide, heterogeneous terrain
Limited connectivity for traditional IT infrastructure
Sensor data siloed away from drone operations
Difficult to decide when and where to send inspection flights
Solution
1. WiseOS – LoRaWAN Integration
WiseOS connected:
- Soil and weather sensors via LoRaWAN into ROS 2 topics
- Edge gateways at strategic locations
- Telemetry stored in InfluxDB
- Single view of field conditions over time
2. MCP ROS2 – Drone Verification
MCP ROS2 exposed tools for:
- Listing zones and their sensor statuses
- Requesting drone inspections over zones with anomalies
- Retrieving recent measurements and flight results
- Correlating sensor readings with aerial imagery
3. Digital Twin – Field Layout
A lean Digital Twin of field zones:
- Basic terrain model and zone boundaries
- Integration with WiseOS telemetry as overlays
- Playback of drone paths and inspection results
- Visual spatial context
Architecture Overview
- LoRaWAN sensors → network server → WiseOS bridge → ROS 2 topics
- Drones → ROS 2 → WiseOS / MCP ROS2
- InfluxDB → historical field condition data
- Digital Twin → field visualisation & scenario exploration
Outcomes
The approach typically leads to:
✅ Targeted Inspections
- Better targeted drone inspections based on sensor data
- Reduced unnecessary flights
✅ Better Understanding
- Improved understanding of field variability over time
- Data-driven agronomic decisions
✅ Scalability
- A scalable pattern that can be extended to new fields and crops
- Reusable infrastructure
✅ Visual Context
- Visual, spatial context for agronomists and operations teams
- Better communication across teams
Next Steps
Extensions include more advanced analytics, additional mobile platforms (ground robots, UGVs), and tighter integration with agronomic decision tools