基於動覺教導之類人型機器人全身動作平衡控制

動覺教導是是一種藉由使用者直接牽引機器人的方式，可簡單與快速地教導機器人動作。先前的研究大部分著重在由人類示範動作來讓機器人學習，本論文提出，使用者在動覺教導的示範方式下教導類人型機器人做全身動作的學習時，機器人能自主地保持平衡。主要目的是如何同時控制機器人維持平衡且使其動作與使用者教導動作相似。因此，本論文提出失衡度與差異性權衡函數給使用者來依其需求調整機器人的平衡度與動作相似程度，並且藉由基因演算法找出最佳近似解，並將此方法應用在機器人全身靜態與動態平衡控制上。我們使用Aldebaran NAO機器人做驗證，實驗結果說明透過動覺教導的示範方式及失衡度與差異性權衡函數，可幫助機器人在使用者的設定下展現出靜態與動態動作相似與平衡

關鍵字

動作示範學習；動覺教導；全身動作模仿；平衡

並列摘要

Teaching humanoid whole-body motion can be done by two ways. One is human-demonstration teaching and the other is kinesthetic teaching. The former is to capture human motion and transfer it to a humanoid robot. The latter is directly pull the robot’s body by a human. Previous work focused on human-demonstration teaching but encountered the correspondence problem which was difficult to be solved. In this thesis, we adopt kinesthetic teaching because it is an intuitive way to teach a robot how to move. This work proposes a method to control the robot’s balance and make the robot motion similar to kinesthetic teaching motion. The proposed method includes the trade-off function of pseudo balance and similarity and a genetic algorithm for solving it. The trade-off function can be adjusted by a user’s preference to choose either better balance or similar motion. The proposed method is applied to both static and dynamic motion. We used an Aldebaran Nao robot to validate the proposed method and the results show that the robot can perform any static pose and kick a ball with one foot standing by kinesthetic teaching.

並列關鍵字

Programming by demonstration ； kinesthetic teaching ； whole-body motion imitation ； balance

參考文獻

[2] S. Calinon, "Robot programming by demonstration," in Springer Handbook of RoboticsAnonymous Springer, 2008, pp. 1371-1394.

[3] A. Bandura and D. C. McClelland, "Social learning theory," 1977.

[6] M. Hersch, F. Guenter, S. Calinon and A. Billard, "Dynamical system modulation for robot learning via kinesthetic demonstrations," IEEE Transactions on Robotics, vol. 24, 2008,pp. 1463-1467

[7] C. L. Nehaniv and K. Dautenhahn, "2 The Correspondence Problem," Imitation in Animals and Artifacts, 2002,pp. 41,.

[12] K. Erbatur, A. Okazaki, K. Obiya, T. Takahashi and A. Kawamura, "A study on the zero moment point measurement for biped walking robots," 2002. 7th International Workshop on Advanced Motion Control, , 2002, pp. 431-436.

國際替代計量

基於動覺教導之類人型機器人全身動作平衡控制

全文下載

主題瀏覽