max/icl-iot-weather
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update

Browse Source
This commit is contained in:
Max 2021-01-06 13:36:09 +00:00
parent 0457118385
commit 17312a5461
1 changed files with 30 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
README.md
View File

@ -3,14 +3,18 @@
#### Reducing water waste by automatically dispensing lower bound water amounts, calculated using a ML Model using present weather data #### Reducing water waste by automatically dispensing lower bound water amounts, calculated using a ML Model using present weather data
## About ## About
About This project aims to integrate hardware IoT sensors (some RPi Zero's with I2C probes) and Machine Learning (Deep-NN) to predict how much water a outdoor garden will need every day.
The ML Model uses the last 24hrs of collected weather data to make a prediction around mow much water needs to be dispensed to keep the plants alive.
This approach hopes to reduce the overall water use by not watering plants in excess.
## Project diagram ## Project diagram
```TODO``` ```TODO```
## File Structure ## File Structure
<!-- <details> --> <details>
<summary>This part is long, click to expand</summary> <summary>This part is long, click to expand</summary>
- `data_collector/` Main data collector and uploader module - `data_collector/` Main data collector and uploader module
@ -50,7 +54,30 @@ About
- `.gitignore` Gitignore file preventing all my API keys from showing up online... - `.gitignore` Gitignore file preventing all my API keys from showing up online...
- `README.md` See `README.md` - `README.md` See `README.md`
<!-- </details> --> </details>
## Installation and running
1. Clone this repo
2. Create the secrets folders
3. Generate a set of firebase and OpenWeather API credentials and place them in the secrets folders, rename accordingly
4. Move the `master` folder to the main Pi
5. Move the `satellite` folder to the sensor Pi
6. Install Docker on all Devices
7. Build the docker images on the relevant architectures (for me everythong was build on armv7)
- `data_colelctor`: `docker build -t siot-weather-collector .`
- `data_processor`: `docker build -t siot-data-processor .`
- `master/irrigator`: `docker build -t master-node .`
- `master`: `docker build -t irrigator .`
- `satellite`: `docker build -t satellite .`
8. *NOTE: The ML Dockerfile is very long and complicated because tensorflow does not play well with a 32 bit arm architeture, if you are building for x86 or arm64, you may need to change the file*
9. Run the docker containers:
- The `data_collector` on the cloud device: `docker run -d --restart always --name siot_weather_collector siot-weather-collector`
- The `data_processor` on the cloud device: `docker run -dp 3535:3535 --restart always --name siot_watering_predictor siot-data-processor`
- The `master` on the master Pi: `docker run -dp 3333:3333 --privileged --restart always master-node`
- The `master/irrigator` on the master Pi: `docker run -d --privileged --restart always irrigator`
- The `satellite` on the sensor Pi: `docker run -d --privileged --restart always satellite`
10. While you are free to use my ML model included in this repo, I suggest you explore the `manual_data_processing` folder and create your own.
11. You will also need at least 24hrs of data before the model can make predictions, so the `siot-weather-collector` image must be started at least 24hrs before the others
## Maintainers and Contributors ## Maintainers and Contributors