臺灣博碩士論文加值系統

姿態辨識是現今人機互動中重要的研究領域，可應用於如醫療復健、虛擬實境和電腦遊戲等不同產業。本研究提出基於慣性感測器之穿戴式裝置應用於虛擬實境之人體動作辨識。首先使用一般消費型之慣性感測器透過演算法將感測器數值轉換成物理角度，再經由量角器進行精度與重複性實驗，檢測本裝置之角度變化準確度。為了讓使用者在動作時不受電線的影響，因此透過無線網路連線方式傳輸感測器數據至電腦，並比較了TCP與UDP兩種不同協定的傳輸速度實驗，完成了準確度與無線傳輸速度實驗後，依此結果證明之可行性，進一步設計新型電路板縮小裝置體積易於穿戴至身上，並比較市售之動態捕捉系統與本研究開發之無線穿戴裝置的差異性。
在相關應用上，本研究採用Unity遊戲引擎開發一套預先設計三種拳路動作與整合本研究裝置的互動遊戲，使用者在虛擬環境中找尋目標物並使用預設拳路擊打，可記錄使用者的分數、出拳種類、次數與力道大小，讓使用者知道自己之身體與遊戲互動情況進而提供良好的虛擬實境沉浸感與遊戲體驗。

Nowadays, human activity recognition is an important research field in human-computer interaction. It can be applied to different industries such as physical rehabilitation, virtual reality and computer games. This study proposes a wearable device based on IMU (Inertial Measurement Units) sensors for human activity recognition in virtual reality. First of all, through algorithm convert the IMU sensor data into physical angle, and then an accuracy and repeatability experiment is performed through the protractor to detect the angle variation accuracy of the device. In order to prevent the user from being affected by the wires during the operation, the sensor data is transmitted to the computer through the Wi-Fi, and the transmission speed experiments of the two different protocols of TCP and UDP are compared. After the finish the accuracy and transmission speed experiments, according to the feasibility of the result, the new circuit board is further designed to reduce the size of the original device make it much easier to setup on the body, and find the difference between the commercially available dynamic capture system and the wireless wearable device developed by the research is compared.
In related applications, this study uses Unity game engine to develop an application of interactive games designed in advance to integrate three kinds of boxing actions and integrate the research device. The user finds the target in the virtual environment and uses the preset punch to hit the target, which can be recorded and used. The player's score, the type of punch, the number of hit times and the magnitude of the force, let the user know the interaction between his body and the game to provide a good virtual reality immersion and gaming experience.

摘 要........i
Abstract......ii
誌 謝.........iii
目 錄..........iv
表目錄...........vi
圖目錄...........vii
第一章 緒論..........1
1.1研究背景與動機.....1
1.2研究目的...........2
1.3文獻探討............2
1.3.1體感操控與技術......2
1.3.2穿戴式裝置與相關技術..2
1.3.3姿態辨識技術..........4
1.4論文架構................4
第二章 研究方法.............5
第三章 系統架構.............10
3.1硬體架構..................10
3.1.1電子元件選用.............10
3.1.2電路板設計...............11
3.1.3外殼設計..................13
3.2軟體架構.....................14
3.2.1程式流程....................14
3.2.2資料傳輸.....................15
3.2.3拳路與力道辨識設計.............19
3.2.4虛擬環境開發...................23
第四章 裝置試驗......................27
4.1角度測試..........................27
4.2準備階段...........................31
4.3操作階段............................34
4.3.1遊戲應用...........................34
4.3.2動畫錄製............................35
第五章 結論...............................38
參考文獻...................................39
Extended Abstract..........................41

[1]VR/AR Global Investment Report & Outlook 2018 Website: https://static1.squarespace.com/static/575e5cd62b8ddeb3fba63f79/t/5aa963d4e2c483ff9d60f3c9/1521050602030/VRVCA_Global+Investment+Report+2018_vF+%28EN%29.pdf
[2]Jamie Shotton, Toby Sharp, Alex Kipman, Andrew Fitzgibbon, Mark Finocchio, Andrew Blake, Mat Cook, and Richard Moore. 2013. Real-time human pose recognition in parts from single depth images. Commun. ACM 56, 1 (January 2013), 116-124. DOI: https://doi.org/10.1145/2398356.2398381
[3]Chenxuan Xi, Jianxin Chen, Chenxue Zhao, Qicheng Pei, and Lizheng Liu. 2018. Real-time Hand Tracking Using Kinect. In Proceedings of the 2nd International Conference on Digital Signal Processing (ICDSP 2018). ACM, New York, NY, USA, 37-42. DOI: https://doi.org/10.1145/3193025.3193056
[4]Preston Tunnell Wilson, Kevin Nguyen, Alyssa Harris, and Betsy Williams. 2014. Walking in place using the Microsoft Kinect to explore a large VE. In Proceedings of the 13th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry (VRCAI '14). ACM, New York, NY, USA, 27-33. DOI: https://doi.org/10.1145/2670473.2670492
[5]J. A. Lozano-Quilis et al., "Virtual reality system for multiple sclerosis rehabilitation using KINECT," 2013 7th International Conference on Pervasive Computing Technologies for Healthcare and Workshops, Venice, 2013, pp. 366-369. DOI: 10.4108/icst.pervasivehealth.2013.252208
[6]S. Rawat, S. Vats and P. Kumar, "Evaluating and exploring the MYO ARMBAND," 2016 International Conference System Modeling & Advancement in Research Trends (SMART), Moradabad, 2016, pp. 115-120. DOI: 10.1109/SYSMART.2016.7894501
[7]Leigh Ellen Potter, Jake Araullo, and Lewis Carter. 2013. The Leap Motion controller: a view on sign language. In Proceedings of the 25th Australian Computer-Human Interaction Conference: Augmentation, Application, Innovation, Collaboration (OzCHI '13), Haifeng Shen, Ross Smith, Jeni Paay, Paul Calder, and Theodor Wyeld (Eds.). ACM, New York, NY, USA, 175-178. DOI: http://dx.doi.org/10.1145/2541016.2541072
[8]Morgan McCullough, Hong Xu, Joel Michelson, Matthew Jackoski, Wyatt Pease, William Cobb, William Kalescky, Joshua Ladd, and Betsy Williams. 2015. Myo arm: swinging to explore a VE. In Proceedings of the ACM SIGGRAPH Symposium on Applied Perception (SAP '15). ACM, New York, NY, USA, 107-113. DOI: https://doi.org/10.1145/2804408.2804416
[9]Khrystyna Vasylevska, Iana Podkosova, and Hannes Kaufmann. 2017. Teaching virtual reality with HTC vive and leap motion. In SIGGRAPH Asia 2017 Symposium on Education (SA '17). ACM, New York, NY, USA, Article 2, 8 pages. DOI: https://doi.org/10.1145/3134368.3139221
[10]曹薈強 林仲志 吳水才． 基於隱馬爾可夫模型的老年人跌倒行為檢測方法研究 《中國生物醫學工程學報》2017年 第2期，文章編號：0258-8021(2017) 02-0165-07 DOI: 10.3969/j.issn.0258-8021.2017.02.006
[11]林本煌，改良型分層隱藏式馬可夫模型於手勢識別之研究，國立台北科技大學電機工程系研究所碩士學位論文，2011。DOI: 10.6841/NTUT.2011.00714
[12]鄧信凱。「應用動態時間校正演算法作為姿態辨識技術之體適能運動系統」。碩士論文，慈濟大學醫學資訊學系碩士班，2014。DOI: https://hdl.handle.net/11296/9due3x。
[13]Alex Sarnoff. Control VR. 2014. Website: https://www.kickstarter.com/projects/ controlvr/control-vr-motion-capture-for-vr-animation-and-mor
[14]Roetenberg, Daniel & Luinge, Henk & Slycke, Per. (2009). Xsens MVN: Full 6DOF human motion tracking using miniature inertial sensors. Xsens Motion Technol. BV Tech.
[15]Hsin Huang.Unity-Runtime-Recorder. 2014. Website: https://github.com/newyellow/
Unity-Runtime-Animation-Recorder