IoT-project

This repository contains files related to university project, that was focused on creation of small IoT system. Plan was to design a system that improve pedestrians safety. To achieve that few components were developed:

• ML people detection script for RaspberryPi
• master-slave protocol
• communication with the server
• server (both frontend and backend) that gathers statistics from crosswalks and visualize them

Prerequisites

Requirements:

• python3.8+
• poetry
• docker (optional)

Before you run this project create .env configuration file for both backend and frontend.

Sample backend configuration (should be placed in ./backend/):

REGISTRY_HOST="localhost:5000"

BACKEND_PORT=8000
BACKEND_HOST="0.0.0.0"
BACKEND_CONTAINER_NAME=iot-backend
BACKEND_IMAGE_NAME=iot-backend

FRONTEND_PORT=9000
FRONTEND_HOST="0.0.0.0"
FRONTEND_CONTAINER_NAME=iot-frontend
FRONTEND_IMAGE_NAME=iot-frontend

OPERATOR_USERNAME=admin
OPERATOR_PASSWORD=secret
RASPBERRY_PI_USERNAME=provider
RASPBERRY_PI_PASSWORD=supersecret

SECRET_KEY=<PUT YOUR SECRET HERE>
ALGORITHM="HS256"

Sample frontend configuration (should be placed in ./frontend/):

REGISTRY_HOST="localhost:5000"

BACKEND_PORT=8000
BACKEND_HOST="0.0.0.0"
BACKEND_CONTAINER_NAME=iot-backend
BACKEND_IMAGE_NAME=iot-backend

FRONTEND_PORT=9000
FRONTEND_HOST="0.0.0.0"
FRONTEND_CONTAINER_NAME=iot-frontend
FRONTEND_IMAGE_NAME=iot-frontend

OPERATOR_USERNAME=admin
OPERATOR_PASSWORD=secret

SECRET_KEY=<PUT YOUR SECRET HERE>
ALGORITHM="HS256"

# use 'http://localhost:${BACKEND_PORT}/api/v1' if you do not run app in docker
BACKEND_URL='http://${BACKEND_CONTAINER_NAME}:${BACKEND_PORT}/api/v1'

Run application

Application can run directly on host or within docker containers.

Running directly on host

Run backend

cd backend
poetry shell
poetry install (requried only if it is the first app launch on the system)
export PYTHONPATH=.
python app/main.py

Run frontend

cd frontend
poetry shell
poetry install (requried only if it is the first app launch on the system)
python index.py

Now you can go to localhost:${BACKEND_PORT}/docs (BACKEND_PORT is set in .env file) to check the API documentation. Frontend is served on localhost:${FRONTEND_PORT}/dash.

Running within containers

Alternatively, you can build docker images with the setup.sh and run both services with the same script.

---

NOTE: setup.sh requires some env variables to be set. So either export them manualy or define them in .env file.

---

$ bash setup.sh  # build images
+ '[' -f .env ']'
+ source .env
++ REGISTRY_HOST=localhost:5000
++ BACKEND_PORT=8000
...
$ bash setup.sh run  # run both containers

You can use docker ps to verify if containers are up.

In order to initialize app database with sample data run init_sample_db.sh.