comp0016-farmbeats

Project intro:

See the project website for more details: http://students.cs.ucl.ac.uk/2021/group28/

Setup guide

The Project consists of 2 standalone programs: Raspberry Pi edge program and Desktop GUI program, located repsectively in src/ directory and gui_code/ directory. Currently there is a single requirements.txt file for both programs.

Edge device

Install requirements with pip install -r requirements.txt. Also required: Grove library, link to be added @Willmish.

Dekstop device

Install requirements with pip install -r requirements.txt. Also required: Pyodbc and MS odbc driver.

To install pyodbc, ODBC header files are required to build pyodbc buildwheel

For ubuntu, package unixodbc-dev is required:

`sudo apt install unixodbc-dev`

Then, install pyodbc:

`pip install pyodbc`

Follow the guide for your OS to install (it should also include an installtion of ODBC header files mentioned above): https://docs.microsoft.com/en-us/sql/connect/odbc/linux-mac/installing-the-microsoft-odbc-driver-for-sql-server?view=sql-server-ver15#ubuntu18

System design

Overall system design (target)

Above is the overall target design of the current system.

Edge device

This diagram shows the flow of messages in the system using Publisher-Subscriber design pattern along with basic annotation of key events in the system:

• sensors being called to collect() data, and sending SensorData objects over pubsub.
• analysers updating its instances of PID class, with new sensor data and getting new output results, attaching to the SensorData object received and sending over pubsub
• data_streamers listening in on messages sent on the database_update.actuator topic, and saves the data to the currently selected DB (default is Azure SQL DB in cloud)
• actuators are listening in on messages sent on the database_update.actuator and pid_update.actuator topics, and the values for the actuators are updated. (fan speed, LED brightness, water pump state, etc)

Most recent progress

Here is a GIF of the Desktop app displaying live data from the Azure DB, while the Rpi is collecting the data and streaming it to the IoT Hub!