WA-LLE: Walking Aid - Limitless Living for the Elderly

This repository contains work that has been done to develop robotic walking aids for the elderly.

This README should always reflect the current state of the repository. If you create a new node, please add it to the Nodes section below with a short description and update the Usage instructions as necessary.

Usage

1. Clone this repo/package into your catkin workspace.

2. Make the package:

cd ~/catkin_ws
catkin_make

3. Source the catkin setup file:

source ./devel/setup.bash

4. Run roscore (required before running any ROS applications):

roscore

5. Run Astra node

roslaunch astra_launch astra.launch

6. Run OpenPose node and adapter

rosrun openpose_ros openpose_ros_node
rosrun hcr_walker hcr_openpose_adapter.py

7. Run gait detection and publisher

rosrun hcr_walker hcr_gait

8. Run IMU and fall detector

rosrun hcr_walker arduino_send_node.py
rosrun hcr_walker hcr_imu.py
rosrun hcr_walker fall_detector.py

9. Run panic button

sudo nodejs src/panicButton/index.js

Nodes

• arduino_send_node : USB serial connection between ROS and Arduino
• hcr_imu : Connects to an IMU and publishes its gyroscope and accelerometer components
• hcr_gait : gait detection node using OpenPose
• fall_data_collector: collects fall data from IMU and gait analysis, and writes them to CSV files
• fall_detector: determines the falling state of the user
• fall_detector_trainer: creates an SVM classification model

IMU requirements

The IMU requires bluez and bluepy. Install them using sudo apt-get install bluez and sudo pip install bluepy.

Fall Detection Requirements

The fall detector nodes need SciPy and sklearn in order to use the SVM (diabled by default). Install them using sudo apt-get install python-scipy python-sklearn.

Contributors

• mdhchan
• sabrinalucy96
• liu09222003
• JunShern
• BaronKhan
• JB515