This is the implementation of an activity recognition system for ESP8266.
The repository contains the following folders:
backendcontains the server implementation in python for predicting the activitysrccontains the code for ESP8266 board
For running this project you need the ESP8266 board with MPU9250 sensor.
For running the program on the board you need Arduino IDE and ArduinoJson library installed. In the src/WebServerUtils.h you need to change the WiFi credentials, optionally you can also change the username and password. You also need to change the serverUrl in src/src.ino to your computer's IP address.
Before running the program you also need to upload the src/data folder to the board. You can do this with Arduino IDE -> Tools -> ESP8266 Sketch Data Upload.
To start the server, run the backend.py. The default server port is 8080.
Feature evaluation and feature extraction was done as described in the article A Study on Human Activity Recognition Using Accelerometer Data from Smartphones by Akram Bayat, Marc Pomplun, Duc A. Tran, which can be found here.