Please use this identifier to cite or link to this item: http://vpet.vtc.edu.hk/dspace/handle/999/563
DC FieldValueLanguage
dc.contributor.authorYe, Lanen_US
dc.contributor.otherHong Kong University of Science and Technology. Industrial Engineering and Engineering Management-
dc.date.accessioned2017-06-27T06:41:28Z-
dc.date.available2017-06-27T06:41:28Z-
dc.date.issued1998-
dc.identifierhttp://hkall.hku.hk/record=b15063640-
dc.identifier.other<a class="btn btn-default" target="_blank" href="http://lbezone.ust.hk/bib/b603850">HKUST Electronic Theses</a>-
dc.identifier.urihttp://hdl.handle.net/999/563-
dc.descriptionx, 102 leaves : ill. ; 30 cmen_US
dc.description.abstractVirtual Environment (VE) has great potential for providing training to users in industry. It has real value for improving the training effectiveness, supporting collaborative training in shared VE, lowering the "transfer barrier" to the real world, and guaranteeing the safety of both human and systems during training. A practical Virtual Reality-based Industrial Training System (VR-ITS) should take advantage of both Virtual Reality (VR) technology and Intelligent Computer-Aided Training (ICAT). While VR technology permits interactive, immersive experience to the trainee, Artificial Intelligent (AI) technology can provide him/her with instructions, by monitoring his/her actions, detecting the errors made by him/her, and evaluating his/her performance. To reach this goal, it is quite important to effectively embed knowledge into an intelligent VR-ITS. The difficulty of developing usable Knowledge Engines (KE) in VR-ITS is not only that the training domain knowledge involved is quite various and complex, but also that the interaction style between the virtual training systems and trainees is quite different from other traditional training approaches. Although researchers have developed some VR-based training systems, there is no effective approach that incorporates the knowledge and experience of human experts into VR-ITS. In this thesis, by analyzing and classifying the specific characteristics of knowledge in the field, a generic architecture of Recursive and Extended Petri Net (REPN)-based and agent-oriented KE for an intelligent VR-ITS was proposed. Furthermore, according to the architecture proposed, a task/action-oriented knowledge elicitation skeleton was depicted, and an hybrid knowledge representation pattern was presented. The reasoning mechanism for the proposed KE was also based on REPN. To demonstrate the proposed KE design and illustrate its effectiveness, KE were integrated into the VR-based CNC milling operation prototyping system by following this methodology. In the end, some experiments were performed in order to examine the effectiveness of the developed system.en_US
dc.publisherHong Kong : Hong Kong University of Science and Technologyen_US
dc.subject.lcshOccupational training -- Data processingen_US
dc.subject.lcshComputer-assisted instructionen_US
dc.subject.lcshVirtual realityen_US
dc.titleAgent-oriented knowledge engines for intelligent virtual reality-based industrial training systemsen_US
dc.typeThesisen_US
dc.identifier.doi10.14711/thesis-b603850-
item.fulltextNo Fulltext-
item.grantfulltextnone-
Appears in Collections:Facilities and Technologies
Show simple item record

Page view(s)

105
checked on Mar 29, 2024

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.