Skip to content

KAIST-NMAIL/SSGMG

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Semi-Supervised Gait Generation With Two Microfluidic Soft Sensors

Dooyoung Kim, Min Kim, Junghan Kwon, Yong-Lae Park, Sungho Jo, "Semi-Supervised Gait Generation With Two Microfluidic Soft Sensors," in IEEE Robotics and Automation Letters, vol. 4, no. 3, pp. 2501-2507, July 2019. / doi: 10.1109/LRA.2019.2907431 / http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8678818&isnumber=8668830

Abstract

Nowadays, the use of deep learning for the calibration of soft wearable sensors has addressed the typical drawbacks of the microfluidic soft sensors, such as hysteresis and nonlinearity. However, previous studies have not yet resolved some of the design constraints such as the sensors are needed to be attached to the joints and many sensors are needed to track the human motion. Moreover, the previous methods also demand an excessive amount of data for sensor calibration which make the system impractical. In this letter, we present a gait motion generating method using only two microfluidic sensors. We select appropriate sensor positions with consideration of the deformation patterns of the lower-limb skins and mutual interference with soft actuators. Moreover, a semi-supervised deep learning model is proposed to reduce the size of calibration data. We evaluated the performance of the proposed model with various walking speeds. From the experiment, the proposed method showed a higher performance with smaller calibration dataset comparing to the other methods that are based on the supervised deep learning.

keywords

IEEE Keywords

Calibration;Sensor phenomena and characterization;Wearable sensors;Capacitive sensors;Data acquisition;Strain

Author Keywords

Wearable robots;soft robot applications;deep learning in robotics and automation},

Usage

SSGMG.ipynb based on google colab

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 71.0%
  • Python 29.0%