NYCU ForeLook

AIoT PD Foot Pressure Sensing Insole

This application is designed to show how to diagnose whether the users have the tendancy of Parkinson's disease using embARC. AIoT PD Foot Pressure Sensing Insole can measure user's gait, and then judge if the user is in high risk of the disease by NN Model. Every components of the device are detachable which makes it superior in mobility and convenience. Also, it can be controlled by Android App. The connection between the device and the Smartphone is based on Wi-Fi.

• Introduction
  • System Architecture
  • Android App
• Hardware and Software Setup
  • Required Hardware
  • Required Software
  • Hardware Connection
• User Manual
  • Before Running This Application
  • Run This Application

Introduction

AIoT PD Foot Pressure Sensing Insole

AIoT PD Foot Pressure Sensing Insole is a smart device which can be used to monitor the high-risk populations of Parkinson's disease. By daily monitoring, users can get the warning before everything get worse and look for the treatment as soon as possible. For PD patients, our project can also be a severity reference.

Demo Link

Click here

System Architecture

Android App

Our device can be controlled by AIoT-PD Android App. You can start/stop the measurement, see your foot pressure destribution and the final result on the App.

Hardware and Software Setup

Required Hardware

• 1 Himax WE-I Plus EVB Endpoint AI Development Board

• 2 Raspberry Pi 3 B

• 2 8-Channel 12-Bit ADC for Raspberry Pi (STM32F030)

• 2 RPi UPSPack V3

• 2 4000mAh Lithium Battery

• 2 Pressure Sensing Insole(Homemade)

• The Pressure Sensing Insole shown below.

  
  • 2 Insole
  • 16 FlexiForce A301 Pressure Sensor (111N)

• The physical picture shown below.

• The structure diagram shown below.

Required Software

• Metaware or ARC GNU Toolset
• embARC Machine Learning Inference Library
• AIoT-PD Android App

Hardware Connection

• Connect WE-I Plus board to Raspberry Pi with USB cable(using UART)
• Connect Raspberry Pi to 8-Channel 12-Bit ADC and RPi UPSPack V3(with Lithium Battery)
• Connect pressure sensing insole to 8-Channel 12-Bit ADC
• Connect Raspberry Pi and mobile phone via Wi-Fi

User Manual

Before Running This Application

• Download source code and AIoT-PD App from github
• Setup hardware connection (The hardware resources are allocated as following table.)

Hardware Resource	Function
FlexiForce A301	Pressure sensor
STM32F030	ADC for Raspberry Pi
RPi UPSPack V3	Portable Power Supply
Raspberry Pi 3	Data Preprocessing, Provide Wifi Connection

Run This Application

To start foot pressure collecting and make disease diagnosis

• Create output_gnu.img

  • Go to the Github of Synopsys and clone or download it.

  • Put the folder foot_project_split_test_merge_0725 in the following path (arc_contest/Synopsys_SDK/User_Project/)

  • Open folder in Visual Studio Code (……/arc_contest/Synopsys_SDK/User_Project/foot_project_split_test_merge_0725)

  • Open Terminal and key-in "make"

  • Open Virtual Machine Ubuntu and go to same project path ({Share Folder…}\arc_contest\Synopsys_SDK\User_Project\foot_project_split_test_merge_0725)

  • Open Terminal and key-in "make flash"

  • Get output_gnu.img

• Open your serial terminal such as Tera-Term on PC, and configure it to right COM port and 115200bps

• Burn output_gnu.img onto WE-I Plus board

• Press reset on WE-I Plus board

• Connect your Raspberry Pi 3 and Mobilephone to same Wi-Fi

• Modify IP on Raspberry Pi 3 and APP to your own

  • Raspberry Pi 3(right foot): right_rpi_server.py

      #Initialize socket parameter of server
      TCP_IP = # Your IP Address(Server)
      TCP_PORT = # Your Port	
    

  • Raspberry Pi 3(left foot): left_rpi_client.py

      TCP_IP = # Your IP Address(Server)
      TCP_PORT = # Your Port
    

  • APP

      fun Myconnect( ){
      	socket = Socket( # Your IP Address , # Your Port )
      }
    

To run Raspberry Pi 3

• Run right_rpi_server.py on right foot Raspberry Pi 3 to create a TCP socket server
• Run left_rpi_client.py on left foot Raspberry Pi 3 to create a client and then connects to the server

To sign up a new ID in AIoT-PD APP

• Open the APP
• Type your Name and Password
• Click register button to register

To run AIoT-PD APP

• Open the APP and sign in
• Press Start button to start data collecting
• Press Stop button to stop data collecting
• After walking for 30 seconds, press Show Result boutton to see the score