臺灣博碩士論文加值系統

本研究建立一助行器使用者之防傾倒系統。透過Openpose之動作捕捉與計算單元之實時擷取使用者姿態，可以計算零力矩點與安全區並以此評估使用者姿態是否穩定。研究中以一左腳為弱腳的下肢輕微失能且駝背患者為例。透過Matlab模擬與實驗發現，若是零力矩點超出安全區，零力矩控制器會改變助行器移動速度以擴大安全區面積來包覆零力矩點。另一方面，透過力敏電阻及荷重元記錄使用者外部受力，提供力學分析所需數據。本研究使用OpenSIM對實驗中蒐集之使用者姿態進行模擬與分析。結果發現經過零力矩控制器補償後的助行器位置可以減輕使用者左腳(弱腳)膝關的關節力矩與相關肌肉的出力。

In this study, a system of real-time fall and overturn prevention for walking-aid users is established. Based on the user’s real-time posture captured with Openpose and calculation unit, the Zero Moment Point (ZMP) and safe region can be calculated and used to assess the stability of current posture. In particular, the gait of a hunchbacked man with mildly disabled lower limb on his left leg (i.e. the weaker leg) is imitated and investigated in this study. Both by simulation with Matlab and experimental verification, it is discovered that, when the ZMP exceeds the safe region, the zero moment controller will adjust the walking-aid's speed to expand the safe region and encompass the ZMP. On the other hand, external forces on the user are recorded with force-sensitive resistors(FSRs) and load cells to provide necessary data for biomechanical analysis. Besides, OpenSIM is used to conduct these simulation and analysis based on user's posture collected during experiments. Simulation results reveal that the zero moment controller can move the walking-aid to proper positions and thus alleviate torque on the left knee joint as well as forces on related muscles.

摘要 i
Extend Abstract ii
致謝 vii
目錄 viii
圖目錄 x
表目錄 xiv
符號表 xv
第一章 緒論 1
第二章 防傾倒系統架構 5
2.1 系統架構 5
2.2 助行器單元 8
2.3 使用者動力學分析 12
第三章 動作捕捉與計算 20
3.1 Kinect v2 20
3.2 人體特徵偵測 22
3.3 飛行時間 27
第四章 零力矩點 30
4.1 使用者與助行器模型 30
4.2 公式推導 33
4.3 Matlab 模擬 38
第五章 實驗與討論 41
5.1 實驗設置 41
5.1.1 實驗設備 41
5.1.2 實驗環境 46
5.1.3 實驗假設 47
5.2 實驗結果與討論 50
第六章 結論與未來展望 65
6.1 結論 65
6.2 未來展望 65
參考文獻 66
附錄A 69
附錄A 87

[1]Scott L et al. Delp. Delp2007.pdf. 54(11):1940–1950, 2007.
[2]Apoorva et al. Rajagopal. Full-Body Musculoskeletal Model for Muscle-Driven Simulation of Human Gait. IEEE Transactions on Biomedical En-gineering, 63(10):2068–2079, 2016.
[3]Microsoft. Microsoft Kinect - Starting to Develop with Kinect,2020. [On-line;accessed 7-July-2020].
[4]Daniil Osokin. Real-time 2d multi-person pose estimation on cpu: Lightweight openpose. arXiv preprint arXiv:1811.12004, 2018.
[5]X.H. Yang. Utilizing depth data and 2d-pose estimation from part affinity fields to develop a tennis training system. Master’s thesis, National Cheng Kung University, 2021.
[6]Larry Li et al. Time-of-flight camera—an introduction. Technical white paper,(SLOA190B), 2014.
[7]T.H. Lu. Dynamic analysis of movement for walking-aid users. Master’s thesis, National Cheng Kung University, 2020.
[8]T.H. Hsieh. The study of shared control applied on a human-and-walking-aid system. Master’s thesis, National Cheng Kung University, 2017.
[9]X.W. Lin. Model-fusion-based precision motion control of linear motors. Mas-ter’s thesis, National Cheng Kung University, 2013.
[10]Wikipedia. Population ageing — Wikipedia, the free encyclopedia, 2020.[Online; accessed 7-July-2020].
[11]cureresearch. Conditions by people affected (prevalence or incidence), 2020.[Online; accessed 7-July-2020].
[12]Hiroyuki et al. Ukida. Human motion capture system using color markers and silhouette. Conference Record - IEEE Instrumentation and Measurement Technology Conference, (April):151–156, 2006.
[13]Takeo Kanade and Masatoshi Okutomi. A stereo matching algorithm with an adaptive window: Theory and experiment. IEEE transactions on pattern analysis and machine intelligence, 16(9):920–932, 1994.
[14]Andrew Forbes, Michael de Oliveira, and Mark R Dennis. Structured light. Nature Photonics, 15(4):253–262, 2021.
[15]T Sakurai. Development of ZMP controls for walking reconstruction and walk-ing assist for the elderly. PhD thesis, Master Thesis, Shibaura Institute of Technology Graduate School, 2012.
[16]Miomir Vukobratovi´c and Branislav Borovac. Zero-moment point—thirty five years of its life. International journal of humanoid robotics, 1(01):157–173, 2004.
[17]T.A.English and M.Kilvington. In vivo records of hip loads using a femoral implant with telemetric output (a prelimary report). Journal of Biomedical Engineering, 1(2):111–115, April 1979.
[18]G.M. Kotzar et al. Telemeterized in vivo hip joint force data: A report on two patients after total hip surgery. Journal of Orthopaedic Research, 9(5):621–633, September 1991.
[19]S.J.G. Taylar et al. An Intrumented Prosthesis F Knee Joint Force Measure-ment in Vivo. The Institution of Electrical Engineers, pages 1–2, 1999.
[20]R. Susmita et al. Modeling of muscle activation from electromyography recordings in patients with cerebral palsy. Applied Sciences (Switzerland), 8(12), 2018.
[21]G. Felipe et al. EMG signal analysis based on fractal dimension for muscle activation detection under exercice protocol. 2016 21st Symposium on Signal Processing, Images and Artificial Vision, STSIVA 2016, pages 13–17, 2016.
[22]OpenSIM. Residual reduction algorithm - opensim documentation - global site, 2020. [Online; accessed 7-July-2020].
[23]P. Brenner. A technical tutorial on the ieee 802.11 protocol. Breezecom Wireless Communications, July 1996.
[24]Texas Instruments. CC2640/CC2650 Bluetooth low energy Software Devel-oper’s Guide (Rev. E). http://www.ti.com/lit/ug/swru393e/swru393e. pdf, March 2018.
[25]C.W. Chiu. The assessment of dynamic torques on joints for walking-aid users. Master’s thesis, National Cheng Kung University, 2018.
[26]et al. Rajagopal. Full-body musculoskeletal model for muscle-driven sim-ulation of human gait. IEEE transactions on biomedical engineering, 63(10):2068–2079, 2016.
[27]D.L. Paolo. Adjustments to zatsiorsky-seluyanov’s segment inertia parame-ters. Journal of biomechanics, 29(9):1223–1230, 1996.
[28]et al. Andrea Corti. A metrological characterization of the kinect v2 time-of-flight camera. Robotics and Autonomous Systems, 75:584–594, 2016.
[29]Miomir Vukobratovic and Davor Juricic. Contribution to the synthesis of biped gait. IEEE Transactions on Biomedical Engineering, (1):1–6, 1969.
[30]Atsuo et al. Takanishi. The realization of dynamic walking by the biped walking robot wl-10 rd. Journal of the Robotics Society of Japan, 3(4):325–336, 1985.
[31]Jiaoyan et al. Tang. Stability control for a walking-chair robot with human in the loop. International Journal of Advanced Robotic Systems, 6(1):3, 2009.
[32]Bo-Rong et al. Yang. Stable posture compensation based on zero-moment point control method for a walking assistance apparatus. In 2018 IEEE In-ternational Conference on Cyborg and Bionic Systems (CBS), pages 286–291. IEEE, 2018.
[33]James William Thomas. Numerical partial differential equations: finite dif-ference methods, volume 22. Springer Science & Business Media, 2013.
[34]R.H. Brown et al. Analysis of algorithms for velocity estimation from dis-crete position versus time data. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 39(1):11–19, 1992.