Skip to main navigation Skip to main content
  • E-Submission

JKSPE : Journal of the Korean Society for Precision Engineering

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS
REGULAR

Poppy Robot의 모션 제어를 위한 디지털 트윈 기반의 머신러닝 및 Openpose의 활용 방법에 관한 연구

A Study on How to Utilize Digital Twin-based Machine Learning and Openpose for Poppy Robot’s Motion Control

Journal of the Korean Society for Precision Engineering 2024;41(5):401-405.
Published online: May 1, 2024

1 국립창원대학교 스마트제조융합협동과정

1 Smart Manufacturing Engineering, Changwon National University

#E-mail: ytcho@changwon.co.kr, TEL: +82-55-213-3608
• Received: February 5, 2024   • Revised: February 16, 2024   • Accepted: February 26, 2024

Copyright © The Korean Society for Precision Engineering

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 34 Views
  • 3 Download
  • 1 Crossref
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • Development of a Machine Learning-Based Predictive Model for the Structural Safety of an Optical Table Air Springs Using Simulation Data
    Hwi Jun Son, Young Tae Cho
    Journal of the Korean Society of Manufacturing Process Engineers.2025; 24(12): 105.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

A Study on How to Utilize Digital Twin-based Machine Learning and Openpose for Poppy Robot’s Motion Control
J. Korean Soc. Precis. Eng.. 2024;41(5):401-405.   Published online May 1, 2024
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
A Study on How to Utilize Digital Twin-based Machine Learning and Openpose for Poppy Robot’s Motion Control
J. Korean Soc. Precis. Eng.. 2024;41(5):401-405.   Published online May 1, 2024
Close

Figure

  • 0
  • 1
  • 2
  • 3
  • 4
  • 5
A Study on How to Utilize Digital Twin-based Machine Learning and Openpose for Poppy Robot’s Motion Control
Image Image Image Image Image Image
Fig. 1 Poppy robot’s wiring diagram overview [9] (Adapted from Ref. 9 on the basis of OA)
Fig. 2 Poppy robot’s digital twin modeling in unity (a) Figure of robot, (b) Parts of poppy robot, (c) Transform of parts of robot, (d) Real robot figure, and (e) A virtual robot posing like a real robot
Fig. 3 Acquiring motor angles in poppy robot's digital twin simulator
Fig. 4 Learning results of regression model for robot motion estimation (a) change in accuracy up to 25,000 epoch, and (b) change in LOSS (MSE) value up to 25,000 epoch
Fig. 5 As shown in (a) and (b), the robot arm is controlled by the regression model as the mouse pointer position changes
Fig. 6 Data on joint points extracted from photos using openpose applied to actual robot
A Study on How to Utilize Digital Twin-based Machine Learning and Openpose for Poppy Robot’s Motion Control
L_shoulder_x L_shoulder_y L_arm_z L_elbow_y Hand_pos_x Hand_pos_y Hand_pos_z
0 -120 -10 -30 -20 -0.3287 1.8475 3.8166
1 -120 -10 -30 -15 -0.3698 1.7979 3.8995
2 -120 -10 -30 -10 -0.4140 1.7420 3.9766
3 -120 -10 -30 -5 -0.4608 1.6800 4.0473
19799 -65 25 25 -10 -2.3436 -0.5908 3.9708
19800 -65 25 25 -5 -2.3134 -0.6913 3.9658
19801 -65 25 30 -90 -2.2400 -0.8850 3.9328
Table 1 Position data of the fingertip per motor’s rotation value

19802 rows X 7 columns