There were no major changes to our design.
Challenges & Mitigation:
- Not enough ML data: This is an ongoing issue that we may not be able to collect enough plant data to sufficiently train the ML model. To solve this, we will start collecting our own data from our plants next week and keep prioritizing setting up sensors.
Progress:
- Web Scraping: Except for pH data, we are done with web scraping that provides plant information essential for growth.
- Plant Identification API: Plant Identification API was set up for identifying plant species.
- Light Sensor: We confirmed our code for light sensor is working.
- Communication between RPi and Web App: We successfully transmitted data from the RPi to the web app, and we’re almost done with transmitting data from the web app to the RPi.
- ML Integration with Web App: RPi captures images, runs the ML model for identification, and sends results to web app.
- RPi Camera Integration with Web App: The camera was successfully integrated with web app.
Next Steps:
- Collect our own data from our plants
- Set up the greenhouse environment for demo
- Develop and test multi-classification ML models
- Finish MQTT setup
- Display web scraped data on web app
- Implement auto-scheduling feature
- Set up all sensor data to be sent to the web app