人型機器人為具有仿人類外型且具有執行仿人類動作能力之機器人裝置。人型機器人為了呈現人類的動作，其架構傾向於被設計為與人類相似之運動關節配置。由於人型機器人之架構通常會有所不同，因此相同人體動作於相異人型機器人上會各自呈現出相近之動作。本論文考慮人體骨架在運動時各關節連動之特性，而提出以「架構對應區塊」為基礎之方法來解決同一動作於相異架構之對應問題。架構對應區塊為一對人體骨架與人型機器人骨架之間共同具有的關節集合，且同一組對應區塊內之關節具有相同或相似的連動特性。區塊內之關節對應情形可進一步分析出1-1、1-n、n-1、m-n四種特定區塊對應方法，且由此方法來產生所需之動作資料。最後我們建構人型機器人運動模型並實做1-1對應方法，將生成之動作資料透過模擬軟體動態呈現人型機器人之動作。實驗結果顯示本論文提出之方法可以有效轉換出與人體動作相似之人型機器人動作。

Humanoid robots are robotic devices which have ability to imitate human motions and have human appearance. In order to perform human motions, their skeletons of the joint deployment tend to be designed to fit human body. However, since the skeleton of each humanoid robot usually vary, so the same human motion performed by different humanoid robots usually slightly differ from each other .In this thesis, considered the joint linkage feature of the human skeleton in movement, we propose an approach which is based on the “structural mapping zone” to solve the mapping problems of the same motion among different skeletons. A structural mapping zone is a set of joints existing both in human skeleton and in the humanoid robot skeleton .And joints in a mapping zone have the same or similar motion feature. The mappings between two zones can be categorized into four zone mapping methods, which are 1-1, 1-n, n-1, and n-m, for generating the motion data. Finally, we constructed a humanoid robot motion model and implemented the 1-1 mapping method to perform the transformed humanoid robot motions in a robot simulator software .The result shows that our method can transform human motions into humanoid robot motion effectively and with high fidelity.

[1] Humanoid animation網站:http://h-anim.org/
[2] Higgins, J.,”Human Movement: An Integrated Approach.” The C. V. Mosby CO.: St. dovis, Mo. ,1997
[3] Rachid Manseur, ROBOT MODELING AND KINEMATICS., Charles River Media, 2006.
[4] Open HRP網站:http://www.is.aist.go.jp/humanoid/openhrp/Japanese/
[5] 晉茂林,機器人學,五南圖書公司,台北,2000.
[6] Pollard N.S., Hodgins J.K., Riley M.J., Atkeson C.G., “Adapting human motion for the control of a humanoid robot”, Robotics and Automation Proceedings. ICRA '02. IEEE International Conference, Vol. 2, pp. 1390-1397, 2002.
[7] Nakazawa A., Nakaoka S., Ikeuchi K., Yokoi K., “Imitating Human Dance Motions through Motion Structure”, Intelligent Robots and Systems IEEE/RSJ International Conference, Vol. 3, pp. 2539-2544, 2002.
[8] Nakaoka S., Nakazawa A., Yokoi K., Ikeuchi, K., “Leg motion primitives for a dancing humanoid robot”, Inst. of Industrial Science., Tokyo University, Japan, May 2004.
[9] Matsui D., Minato T., MacDorman K.F., Ishiguro H., “Generating Natural Motion in an Android by Mapping Human Motion”, Intelligent Robots and Systems IEEE/RSJ International Conference, pp. 3301-3308, 2005.
[10] Kaneko K., Kanehiro F., Kajita S., Hirukawa H., Kawasaki T., Hirata M., Akachi K., Isozumi T., “Humanoid robot HRP-2”, Robotics and Automation Proceedings, ICRA '04. 2004 IEEE International Conference, Vol. 2, pp. 1083-1090, 2004.
[11] Omer A.M.M., Yu Ogura, Kondo H., Morishima A., Carbone G., Ceccarelli M., Hun-ok Lim, Takanishi A., “Development of A Humanoid Robot Having 2-DOF Waist and 2-DOF Trunk”, Humanoid Robots 5th IEEE-RAS International Conference, pp. 333-338, 2005.
[12] 台灣地區勞工人體計測資料庫網站:http://www.iosh.gov.tw/ergo.htm
[13] Sing Bing Kang, Ikeuchi K., “Toward automatic robot instruction from perception-mapping human grasps to manipulator grasps”, Robotics and Automation IEEE Transactions, Vol. 13, Issue 1, pp 81-95, 1997.
[14] 刘杰, 孙汉旭, 张玉茹, “人手到灵巧手指尖运动映射的实现”, 北京机械工业自动化所机器人中心, 北京邮电大学自动化学院, 北京航空航天大学机器人研究所, 2003.
[15] Webots Reference Manual release 5.1.9, Cyberbotics Ltd.