請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/79088
標題: | 建立一沉浸感與即時性導向之擴增實境銑削模擬訓練系統 Building an Immersive and Real-Time Oriented Augmented Reality System for Milling Simulation and Training |
作者: | Yu-Hsi Chen 陳昱熙 |
指導教授: | 陳湘鳳(Shana Smith) |
關鍵字: | 擴增實境,銑床訓練,遮蔽處理,深度修補,RealSense?,HTC Vive,Leap Motion, Augmented Reality,Milling Machine Training,Occlusion,Depth Recovery,RealSense?,HTC Vive,Leap Motion, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 隨著電腦繪圖技術與硬體處理速度的進步,擴增實境(Augmented Reality, AR)已經廣為人知。其運用不只是在遊戲方面,許多教學或是模擬方面都可以藉由擴增實境添加視覺上的輔助,提高學習或是規劃的效率。本研究希望利用擴增實境,解決傳統工具機技術因為人才的流失與老化,使得經驗傳承的加工技術失傳的問題。同時,也希望藉由擴增實境的輔助,降低教學時的限制與危險,並以不同於傳統一對多的教學方式,提升過程中的學習效率。
本研究建立了一個擬真的擴增實境銑削教學模擬系統。其中,使用者能以自己的雙手與虛擬銑床互動,就如同現實世界中以雙手操控銑床。在顯示虛擬物體時,也運用深度修補與遮蔽處理來提升視覺與互動上的沉浸感。為此,本研究結合了Intel® RealSense™ R200攝像機獲取室內場景的彩色和深度資訊;Leap Motion來追蹤使用者的手部姿態;HTC Vive頭戴式顯示器結合真實場景和虛擬物體。整體專案中,而為了讓使用者在教學模擬的過程中不會因為操作問題而分心,擴增實境中的操作與互動講究與真實互動情形盡可能的相同。同時,擴增實境與虛擬實境的使用上最大的問題就是因顯示延遲引發之暈動症。本研究運用GPU資源輔助運算以減低延遲在成暈動症的機會。 最後,使用者測試結果顯示,本研究所提出的AR訓練系統是直覺、流暢且具有樂趣的,互動方面經過視覺效果的處理後也更加真實容易。 With the improvements made in fields of computer graphics and computer hardware, Augmented Reality (AR) is now a popular technology. Not only can it be used in computer games, but also in the fields of education and simulation. With its virtual aids, learning and planning efficiency can be improved easily. In this study, we hope to utilize the advantages of augmented reality and solve existing difficulties like uneven resource and the safety issue in traditional machine tool training, and preserve the valuable techniques of decreasing and aging technicians. In this study, we created an AR-based milling machine training system. Users can operate a virtual milling machine by using their natural behaviors and look through the AR environment with a depth-recovered occlusion effect. Intel® RealSense™ Camera R200 was used to get the depth information of the indoor scenes. Leap Motion was used to get user’s hand gestures. HTC Vive head-mounted display was used to merge real scene images and virtual images. A calibration method was also developed to solve the dynamic occlusion problem in real time. GPU computing is used to reduce latency. Finally, user test results show that the training effect of the developed AR system is better than the traditional education video. The user test also shows that the system is intuitive, smooth, and interesting. In this study, the feasibility of the AR system was validated and could be used as a reference for future AR-based training and simulation for machine tools. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/79088 |
DOI: | 10.6342/NTU201802939 |
全文授權: | 有償授權 |
電子全文公開日期: | 2023-08-21 |
顯示於系所單位: | 機械工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-R05522629-1.pdf 目前未授權公開取用 | 5.94 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。