...
Here is a top view of ecosystem:
Also with added AI services, Edge Node may act as distributed AI node with federated learning capabilities.
Here is how the system components are deployed and updated:
...
3. Update software stack on the Edge Node by building container images, uploading them to private registry server
(or public registry or available via DockerHub, Amazon ECR, Google GCR, Microsoft ACR, etc.)
...
- Clone SmartAg-foundation repo. And change directory to tools.
- Build BSP image, write to SD card. Insert 16GB+ SD card in drive and run Edge Node
sudo ./write_bsp_image example-edge-node
- Run Edge Server
./start_local_edgeserver
- Solder sensor and USB converter (described in p.6 from Milestone 2: Outdoor single sensor)
- Create new example stack. By default new stack will contain a workable solution with telegraf and exec plugin for S-Soil MTEC-02B.
./create_stack ssoil-mtec-02b-single
- Insert sensor to USB port of Edge Node
- Build stack to get retrieve data (Total Dissolved Solids (TDS), Temperature, Volumetric Water Content (VWC)) from S-Soil MTEC-02B, deploy
./build_and_deploy ssoil-mtec-02b-single example-edge-node
- Open Grafana from Edge Server http://localhost:3000/, login, explore and display telemetry data
Once InfluxDB data source needs to be configured as below:
URL - http://influxdb:8086
Database - telegraf
User - user
Password - password
And press "Save and Test" - Open Explore tab (http://localhost:3000/explore)
You should see new measurement mtec-02b-sensor and fields for it (tds, temperature, vwc) - (Optional) Extend service with new logic to read soil salinity (it's covered in hands on demo video at the beginning).
Deploy and view new data on a dashboard (salinity should be available)
Check - check how to do it in this part of video https://youtu.be/mCqbzVYJyVs?t=1306
Deploy and view new data on a dashboard (salinity should be available)