(Side project not intended to be used in production applications)
Node.js module to predict indoor location using machine learning and wifi information.
Inspired by the Python module whereami by kootenpv
Built using node-wifi and random-forest-classifier
npm install whereami.js
In each room you'd like to use, record data by using the command whereamijs learn <room>
.
Example:
whereamijs learn kitchen // or -l kitchen
The output of running this command will be a JSON file saved in a whereamijs-data
folder with the wifi info.
This command takes a few seconds to get wifi data and save it
After recording training data with the learn
command, run the predict
command to get the room predicted from live data.
whereamijs predict // or -p
You can list the rooms you already have data for, using the rooms
or -r
command.
whereamijs rooms // or -r
Here are some ideas of what it could be used for:
- IoT: Turn on/off lights based on which room you're in.
- Pause TV when leaving a room.
- Block notifications when in the bedroom.
Clone this repo, cd
into it and run node server.js learn <room>
or node.js server predict
.