Doctoral Dissertations

Md. Al-Amin

Keywords and Phrases

Convolutional Neural Network; Cyber Manufacturing; Deep Learning; Skeletal Data; Workforce Training

Abstract

“Production innovations are occurring faster than ever leading conventional production systems towards cyber manufacturing. Manufacturing workers thus need to frequently learn new methods and skills. In fast-changing, largely uncertain production systems, manufacturers with the ability to comprehend workers’ behavior and assess their operational performance in near real-time will achieve better performance than peers. Recognizing worker actions in near real-time while performing the assembly can serve this purpose. However, reliably recognizing the assembly actions performed by the workers is challenging, because the actions for assembly are complex and workers are not only heterogeneous but sensitive to the variation of the work environment. Moreover, assembly data captured by one category of sensors lack reliability. Therefore, this study proposed multimodal sensor data-based action recognition (ActRgn) with novel fusion methods with improved recognition performance. Myo-armband, a wearable sensor was used to capture both Inertial Measurement Unit (IMU) and electromyography (EMG) signals of assembly workers. Microsoft Kinect, a vision-based sensor, was also utilized to track predefined skeleton joints. Captured sensor data was used to train Convolutional Neural Network (CNN) models. Different techniques are proposed to adopt these trained models to the individual new worker to address the issues of between-subject heterogeneity and within-subject variance. Then, various fusion methods were implemented to integrate the prediction results of independent models to yield the final prediction with improved accuracy. One operation in assembling a Bukito 3D printer, which is composed of seven actions, is used to demonstrate the implementation and assessment of the proposed method. Results from the study have demonstrated that the proposed approach effectively improves the prediction accuracy at both the activity level and the subject level. The work of this study builds a foundation to assess and assist the assembly workers in a cyber manufacturing scenario”--Abstract, page iii.

Advisor(s)

Dagli, Cihan H., 1949-
Qin, Ruwen

Committee Member(s)

Allada, Venkat
Corns, Steven
Leu, M. C. (Ming-Chuan)

Department(s)

Engineering Management and Systems Engineering

Degree Name

Ph. D. in Systems Engineering

Publisher

Missouri University of Science and Technology

Publication Date

Fall 2021

Pagination

xiii, 105 pages

Note about bibliography

Includes bibliographic references (pages 99-104).

Rights

© 2021 Md. Al-Amin, All rights reserved.

Document Type

Dissertation - Open Access

File Type

text

Language

English

Thesis Number

T 11935

Download

Share

 
COinS