Skip to content

Device_compatibability.md

Latest commit

 

History

History
122 lines (98 loc) · 5.68 KB

Device_compatibability.md

File metadata and controls

122 lines (98 loc) · 5.68 KB
Raw

HIL Toolkit: Acquisition, estimation and optimization of human in the loop experiments

Toolkit for human in the loop optimization

Table of contents

Requirements for the toolkit

General requirements

  • Python 3.10 or higher.
  • Labstreaminglayer - install here.
  • pylsl - install by running pip install pylsl.

Metabolic cost.

  • Cosmed or other metabolic cart device.
  • Setup metabolic cost acquisition code.

ECG

  • Polar H10
  • Bluetooth capible laptop for ECG.

Foot pressure.

  • Pressure SDK.

Installation

Complete python HIL toolkit install using pip in main directory.

pip install -e .

HIL optimization.

  • First setup the device as mentioned in the device setup section.
  • Setup the optimization problem in the config/<optimization>.yml file. An example of the optimization problem is provided in the config/ECG_config.yml file.
  • Setup optimization using HIL toolkit. Example of this is provided in the scripts/ECG_optimization.py file.
  • Run the optimization script. python scripts/ECG_optimization.py

Device setup

Cost acquisition

Custom sensor setup

  • Since the most physiological sensors are not open source, we have to use provided an acquisition script for the sensors.
  • Example of this script is provided in the scripts/cost_acq.py folder.
  • This script will help setup the cost acquisition device and send the data to the optimization pc.

ECG device setup

For setting up the new polar H10 ( ECG sensor )

  • Please turn on the computer bluetooth connection.
  • Run the following script to find all the available POLAR sensor scripts python scripts/search_polar.py. This script will search for polar sensors in the bluetooth range and save the BLE information in the config/polar.yml file. And return success or failure.
  • Run the script to collect data from the polar and send it. python scripts/collect_polar.py

Exoskeleton/prosthesis

Initial setup

Follow these steps to the setup the exoskeleton device.

  • Connect the device to the network or use a ethernet cable to connect the device to the computer.
  • Check the IP address of the device. The IP address can be found in the control panel of the device or use command window, type ipconfig or ifconfig and find the IP address of the device.
  • Check if the device is connected using ping <IP address of the device>. If the device is connected it will return the response from the device.
  • Create a server in the optimization pc for testing the communication.
  • If control system is in matlab/simulink use the matlab script provided here to test the communication. comms/communication_test.m
  • If control system is in python use the python script provided here to test the communication. comms/communication_test.py
  • If the communication is successful, the device is ready to be used for the optimization.

Adding Exoskeleton/prosthesis in optimization

  • Add the device the following lines in the config/<optimization>.yml
Exoskeleton: 
  port: 5555
  ip: "192.168.0.10"

Cost Estimation

Warning The estimation is not fully tested. Please use with caution.

Metabolic cost estimation.

Please refer the following paper for the details of the estimation.

Reference

Kantharaju, Prakyath, and Myunghee Kim. "Phase-Plane Based Model-Free Estimation of Steady-State Metabolic Cost." IEEE Access 10 (2022): 97642-97650.

Estimation in scripts for offline.

  • Run the following script to estimate the metabolic cost. python scripts/estimate_metabolic_cost.py
  • This script will estimate the estimate the metabolic cost data in the provided in the data/met_data.npy.
  • The scrip will generate the estimation and the also video as shown in the following figure. Metabolic cost estimation
  • It will also make a video of the estimation and the actual data. The video will be saved in the scripts/Results/videos/ folder. Metabolic cost estimation

Estimation in notebooks.

  • Run the following notebook . notebooks/preprocessing_cosmed.ipynb to convert the raw data to the metabolic cost data.
  • To perform the estimation run the following notebook. notebooks/estimation_cosmed.ipynb