國立虎尾科技大學 |

Communication Modality and Activity Recognition in Smart Manufacturing.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Communication Modality and Activity Recognition in Smart Manufacturing./
Author:	Chen, Haodong.
Description:	1 online resource (216 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
Subject:	Mechanical engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9798383099278

Communication Modality and Activity Recognition in Smart Manufacturing.
Chen, Haodong.

Communication Modality and Activity Recognition in Smart Manufacturing. - 1 online resource (216 pages)

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

Thesis (Ph.D.)--Missouri University of Science and Technology, 2024.

Includes bibliographical references

Advancements in sensors, computational intelligence, and big data have been rapidly transforming and revolutionizing the manufacturing sector, leading to robot-rich and digitally connected factories. To facilitate such a transformative process, workforce training and coordination between the human workforce and robots have remained a major challenge. This study aims to create intelligent workforce training systems for human-computer interaction and human-robot collaboration. It proposes the use of dynamic gestures and speech commands for seamless communication between a worker and a robot, leveraging Convolutional Neural Networks (CNN) and multiple threading for recognition and integration. Additionally, a fine-grained activity recognition model has been crafted using transfer learning and Recurrent Neural Networks (RNN) to accurately recognize and predict assembly operations. To further enhance workforce training, a gaze-oriented workforce training system using CNN and Long Short-Term Memory (LSTM) for step recognition, video transformers for repetitive action counting, and gaze estimation for real-time instructional guidance has been developed. The effectiveness of the proposed comprehensive workforce training methods and systems is underscored by promising outcomes and superior performance metrics, including gesture recognition accuracy above 98.00%, speech recognition accuracy over 95.00%, fine-grained activity recognition accuracy at 99.98%, and a 59.98% Off-By-One-Accuracy (OBOA) in repetitive action counting, surpassing current benchmarks. The real-time demonstrations have convincingly illustrated the practicality and efficiency of the proposed methods in human-robot collaboration and advancing workforce training.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798383099278Subjects--Topical Terms:

557493
Mechanical engineering.
Subjects--Index Terms:

Computer visionIndex Terms--Genre/Form:

554714
Electronic books.

Communication Modality and Activity Recognition in Smart Manufacturing.
LDR:03144ntm a22003977 4500 001 1150155
005 20241022111601.5
006 m o d
007 cr bn ---uuuuu
008 250605s2024 xx obm 000 0 eng d
020 $a 9798383099278
035 $a (MiAaPQ)AAI31140220
035 $a AAI31140220
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Chen, Haodong. $3 1476590
245 1 0 $a Communication Modality and Activity Recognition in Smart Manufacturing.
264 0 $c 2024
300 $a 1 online resource (216 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
500 $a Advisor: Leu, Ming C.
502 $a Thesis (Ph.D.)--Missouri University of Science and Technology, 2024.
504 $a Includes bibliographical references
520 $a Advancements in sensors, computational intelligence, and big data have been rapidly transforming and revolutionizing the manufacturing sector, leading to robot-rich and digitally connected factories. To facilitate such a transformative process, workforce training and coordination between the human workforce and robots have remained a major challenge. This study aims to create intelligent workforce training systems for human-computer interaction and human-robot collaboration. It proposes the use of dynamic gestures and speech commands for seamless communication between a worker and a robot, leveraging Convolutional Neural Networks (CNN) and multiple threading for recognition and integration. Additionally, a fine-grained activity recognition model has been crafted using transfer learning and Recurrent Neural Networks (RNN) to accurately recognize and predict assembly operations. To further enhance workforce training, a gaze-oriented workforce training system using CNN and Long Short-Term Memory (LSTM) for step recognition, video transformers for repetitive action counting, and gaze estimation for real-time instructional guidance has been developed. The effectiveness of the proposed comprehensive workforce training methods and systems is underscored by promising outcomes and superior performance metrics, including gesture recognition accuracy above 98.00%, speech recognition accuracy over 95.00%, fine-grained activity recognition accuracy at 99.98%, and a 59.98% Off-By-One-Accuracy (OBOA) in repetitive action counting, surpassing current benchmarks. The real-time demonstrations have convincingly illustrated the practicality and efficiency of the proposed methods in human-robot collaboration and advancing workforce training.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Mechanical engineering. $3 557493
650 4 $a Computer science. $3 573171
653 $a Computer vision
653 $a Industrial AI
653 $a Machine learning
653 $a Smart manufacturing
653 $a Long Short-Term Memory
655 7 $a Electronic books. $2 local $3 554714
690 $a 0548
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Missouri University of Science and Technology. $b Mechanical Engineering. $3 1213808
773 0 $t Dissertations Abstracts International $g 85-12B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31140220 $z click for full text (PQDT)