Walkable

An open & crowd sourced walking assistant for safe, fast & resourceful walks around your place.

Comment

Walkable

An open source & crowd sourced walking navigation for safe, fast & resourceful walks around your place.

Uses completely anonymous location & reporting data from people around the town & machine learning to suggest safe, fast and enjoyable trips around town. Reports and visualizes insights from user data to public safety to improve security in towns

Front-End of the application was built using react-native, react-native-maps, google maps API. Back-end was made using phython flask.

Uses

For Civillian

  • Find safe routes to your destination.
  • Find out details about routes you walk.
  • A heatmap to avoid over crowded areas or find popular areas.
  • Find events and activities near your walking routes, join if you are adventurous or avoid if you don't want to be bothered

For Public Safety

  • Analyze popular routes by users and ensure high security by focusing on those areas
  • Identify and analyze security concerns in crowded areas
  • Get rich insights from crowd sourced data over the cloud using our dashboard
  • Give and receive general community feedback on paths that people take every day

Open Sourcing the platform

We strongly believe that this tool needs to be open sourced in order to maintain transparency, credibility and allow government and city officials to maintain their own version of the database and data analytics to empower their communities and economic development.

How to run the backend

The backend is implemented in Python Flask and contains the rest endpoints for the front end to access. The backend is deployed on Heroku.

Instructions

First, you'll need to clone the repo.

$ git clone https://github.com/metemorris/hackillinois2018.git

Second, let's download pip, virtualenv, foreman, and the heroku Ruby gem.

$ sudo easy_install pip
$ sudo pip install virtualenv
$ sudo gem install foreman heroku

Now, you can setup an isolated environment with virtualenv.

$ virtualenv --no-site-packages env
$ source env/bin/activate

Installing Packages

Gevent

To use gevent, we'll need to install libevent for the gevent production server. If you're operating on a Linux OS, you can apt-get install libevent-dev. If you're using Mac OS X, consider installing the homebrew package manager, and run the following command:

$ brew install libevent

If you're using Mac OS X, you can also install libevent through a DMG available on Rudix.

Without Gevent

If you'd rather use gunicorn without gevent, you just need to edit the Procfile and requirements.txt.

First, edit the Procfile to look the following:

web: gunicorn -w 4 -b "0.0.0.0:$PORT" app:app

Second, remove gevent from the requirements.txt file.

pip

Then, let's get the requirements installed in your isolated test environment.

$ pip install -r requirements.txt

Running Your Application

Now, you can run the application locally.

$ foreman start

You can also specify what port you'd prefer to use.

$ foreman start -p 5555

Deploying to Heroku

$ git push heroku master
$ heroku scale web=1

Finally, we can make sure the application is up and running.

$ heroku ps

Now, we can view the application in our web browser.

$ heroku open

And, to deactivate virtualenv (once you've finished coding), you simply run the following command:

$ deactivate

Rest endpoints

After you've got your application up and running, there a couple next steps you should consider following

/update/

POST request

It takes in:

  • latitude (required)
  • longitude (required)
  • user id (required)

This updates the user's location in the SQL database and also on firebase for real time updates.

/updateIncidents/

POST request

It takes in:

  • latitude (required)
  • longitude (required)
  • type (required)

This updates the incident's location, along with the type in the SQL database and also on firebase for real time updates.

/get/heatmap/

POST request

It takes in:

  • latitude (required)
  • longitude (required)

This gives the heatmap near the location.

/get/traffic/

POST request

It takes in:

  • latitude (required)
  • longitude (required)

This gives the traffic around a certain location.

/get/incident/

POST request

It takes in:

  • latitude (required)
  • longitude (required)

This gives the incident around a certain location.

Testing | Simulating users

To simulate users, you can run simulation.py.

Simulation.py contains the range of latitude/longitude you want your 'users' to be simulated within. You can also specify how much distance the users should move, and also specify the duration in which they will trigger an incident. Along with the amount of users that can be simulated, even the randomness of an incident generation can be adjusted.

Share this project:

Updates