接觸力及拍擊動作於人體組織之分析於機器人按摩療程之應用

本篇論文著重在健康照護機器人於按摩療程方面之應用，主要分為如下兩個部份: 第一部分，我們探討了機器手臂之按摩力量分析，與傳統按摩器具比較起來，究竟要怎樣才能做得比這些按摩器具更好，甚至像是真正的按摩師一樣呢？要解決這個問題，有非常多地方值得探討。首先，機器人是否能估測按摩時所需的力道？由於人體各個部位的組織成份不盡相同，有的部位肌肉厚實、有的部位皮下就是骨頭。若僅僅是用相同的力道去按壓各個部位，不僅無法得到按摩的療效，甚至有可能使對象受傷。而真正的按摩師則是透過按壓各部位時的手感、對象的反應等等當作一個回饋來修正按壓的力道。所以，透過適當的回饋機制使機器人在每次按壓當中能夠評估力道是非常重要的。本論文透過在我們台大NTU-iCeiRA實驗室發展之七軸機器手臂發展自適應式的接觸力量化與補償分析，藉由機器手臂末端之力感測器與手臂運動學之概念，由力量-深度圖之斜率可以辨識出人類肌肉之軟硬性質，並以此做為肌肉之硬度指標; 由此硬度指標，機器手臂可以根據肌肉之軟硬性質不同而施予2Hz~4Hz的按壓頻率，以及0-15kgw下的不同之力道。第二部分，為了擴充機器人按摩手法之多樣性，本篇論文加入了拍擊動作之研究，此亦為機器人按摩研究領域中尚未探討過之手法。與一般揉、捏、壓等手法相異，拍擊動作在探討力量之餘，也需要考慮拍擊前的手臂末端速度、加速度等資訊。因此，除了利用阻抗控制以確保實驗對象的安全之外，我們也利用線上軌跡產生器給予機器人運動的速度、加速度上限，並由此改變拍擊之力道；在拍擊位置方面我們則是發展了卡氏座標下的教點功能以因應人類體型之相異度，以及”虛擬目標點”之功能產生較合適的拍擊力。實驗驗證方面，我們利用雙手臂六軸類人形機器人實現拍擊動作，藉由物理學上之衝量動量理論以及波形相似度分析，比較了人與機器人之間拍擊狀況之量化比較，得到機器人與治療師拍擊波形約有65%的相似度，且可以提供1Hz~2Hz的動作頻率，說明了機器人拍擊按摩之可行性。此外延伸了拍擊動作，藉由換上不同之手臂末端器具(例:拍痰杯)並以拍擊聲音之頻率分析來觀察機器手臂與人類拍擊時的相異度，以作為另一種驗證之方式。

關鍵字

照護機器人；機器手臂；拍擊按摩；接觸力分析；肌肉組織性質

並列摘要

The thesis focuses on the robot control and the massage therapy applications, we divide the content into two parts as followings: In Part I, we discuss the massage force analysis. To make an improvement of the conventional massage tool, the robot needs to know how much force it should exert, and the force should depend on the muscle properties and the human feelings. Thus, we propose a quantization method that the contact force can be adaptively generated by the 7-DoF manipulator on human muscles. The mechanical properties of different muscles can be analyzed under large deformations like the massage by the force-indentation plot. Utilizing stiffness index (SI) of the human muscles, the manipulator can adjust the exerting forces according to the states of muscles during a therapeutic procedure. In part II, we focus on the massage tapping motion research. We utilize the online trajectory generator based on the impedance control to ensure the safety of human and use different velocity and acceleration value as the limit to generate different force. We implement tapping motion with the anthropomorphic dual arm robot, and the Cartesian space teach function to record position as the control input. By the waveform similarity analysis and the impulse-momentum theory, the feasibility of the percussive massage therapy based on robotic tapping motion is verified by the experiments. Besides, we extend the percussive massage to a chest physiotherapy with different end-effector. In conclusion, we propose the robotic acupressure that can exert 0~15kgw with the motion frequency 2~4Hz, and robotic percussive massage that can exert 0~10kgw with the motion frequency 1Hz~2Hz.

並列關鍵字

health care robot ； percussive massage ； contact force analysis ； human tissue ； muscle tissue properties

參考文獻

[1] R. C. Luo, S. Y. Chen, and K. C. Yeh, "Human body trajectory generation using point cloud data for robotics massage applications," in 2014 IEEE International Conference on Robotics and Automation (ICRA), 2014, pp. 5612-5617.

[4] H. Ishii, H. Koga, Y. Obokawa, J. Solis, A. Takanishi, and A. Katsumata, "Development and experimental evaluation of oral rehabilitation robot that provides maxillofacial massage to patients with oral disorders," The International Journal of Robotics Research, vol. 28, pp. 1228-1239, 2009.

[5] R. C. Luo and C. C. Chang, "Electromyographic evaluation of therapeutic massage effect using multi-finger robot hand," in Robotics and Automation (ICRA), 2011 IEEE International Conference on, 2011, pp. 2431-2436.

[9] K. Mouri, K. Terashima, P. Minyong, H. Kitagawa, and T. Miyoshi, "Identification and hybrid impedance control of human skin muscle by multi-fingered robot hand," in 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2007, pp. 2895-2900.

[11] J. Wang and Y. Li, "Massaging human feet by a redundant manipulator equipped with a tactile sensor," in 2010 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, 2010, pp. 7-12.

國際替代計量

接觸力及拍擊動作於人體組織之分析於機器人按摩療程之應用

全文下載

主題瀏覽