人形機器人的人機介面設計與運動控制之研究

The Design of Human-Machine-Interface for the Motion Control of Humanoid Robots

黃俊豪(Chun-Hao Huang)

指導教授 ： 黃志良

淡江大學/工學院/電機工程學系碩士班/碩士(2010年)

https://doi.org/10.6846/TKU.2010.00135

摘要

本論文以小型單板電腦Roboard-100與Microsoft Visual Studio 2008軟體開發具有身高58公分，重量3.5公斤，及全身共22個自由度之人形機器人的人機介面(Human-Machine-Interface, HMI)，並利用遠端電腦登入此嵌入微處理器開啟所設計的人機介面，執行所撰寫的相關程式以進行人形機器人之操控。
所謂的人形機器人即是它與人類的體型相似，而且能在很多情形下幫助人類完成相關任務，因此它必須具備與人類相似的運動能力，例如，走路、轉彎、跨越障礙物、上下樓梯等。本論文首先定義人形機器人的基本動作，例如，微步伐、小步伐、中步伐、大步伐之直走、左右之橫移動作、小角度游移腳之旋轉、小至大角度之左右旋轉、及左右手部之運動。就我們所知的人機介面中，乃是根據逆運動學推導相關馬達所需轉動之角度，下命令於相關馬達以獲得所想要的姿態，然而這種方式相當耗費時間，不適合於即時的任務。本論文所設計的介面將採取不同的方式來創建機器人的動作，首先以卸力的功能取消馬達之Holding力矩，再任意地轉動機器人的自由度以獲得想要的姿態，接著恢復馬達的Holding力矩，並經由回授的功能紀錄與儲存此姿態的相關馬達之角度。
最後以兩個例子：一為人形機器人跨越障礙物，另一為人形機器人至目標點抓取物件，以驗證所建議的方法之有效性及可行性。首先調查與紀錄機器人在跨越障礙物時，在不同的步態當中，若要維持身體平衡，需要修正那幾個自由度，並整合到介面中，接著進行跨越障礙物連續動作之實驗，觀察相關實驗是否達到所規劃的目標，若沒達成則進行適當之修正，以獲得更好的結果。同理，人形機器人至目標點抓取物件之任務，直到可獲得滿意的結果。然而本論文所建議之方法屬於前饋式的，對於超出預期以外的不平衡或不精確的導引結果是無法補償的。

關鍵字

人型機器人 ； 人機介面 ； 運動控制 ； 跨障礙 ； 夾取物件

並列摘要

Both small-size single board computer Roboard-100 and Microsoft Visual Studio 2008 Software are applied to develop the HMI (Human-Machine-Interface) for small-size humanoid robot. This HMI in the humanoid robot (HR) with height 58 cm, weight 3.5 kg, and 22 degrees of freedom is executed by the remote way from another personal computer. The architecture and function of a humanoid robot is similar with the human; in many circumstances it can help human to finish the assigned tasks. Based on these requirements, the HR must possess the capacity for human normal exercise, e.g., walking, turning, stepping over an obstacle, climbing up and down the stairs. In the beginning, the fundamental motions of the HR, e.g., walking in various step lengths, left and right movements, left and right turning in various degrees, and left and right movements of two hands, are defined and illustrated. In general, the movement commands in the HMI are according to the values derived by the inverse kinematics; it is time consuming and not suitable for the tasks in real time. On the other hand, we have easier and faster way to finish the corresponding tasks. First, the holding torque of the HR is relaxed. After the posture adjustment of the HR by hand, it can be in various postures. In this situation, the holding toques of the HR are turn on, and then the corresponding angles of motor for the specific posture are feedback to Roboard-100 and then recorded. Based on the synthesis of walking gaits for a specific task, the balance is considered in very DDP and SSP. The possible destination is navigated by the proposed HMI with appropriate program. Finally, two experimental cases: one is the stepping over an obstacle, the other is the walking and grasping an object, are arranged to confirm the effectiveness and efficiency of the proposed HMI. Because the proposed methodology is a forward type, the unbalance over the expectance or the inaccurate navigation caused by the uncertainty can not compensate.

並列關鍵字

Humanoid Robot ； Human-Machine-Interface ； Motion Control ； Stepping Over Obstacle ； Object Grasping

參考文獻

[7] 楊玉婷，自主人形機器人的避障與視覺為基射門之設計與實現，淡江大學電機工程學系碩士班碩士論文，2007。

[1] K. Loffler, M. Gienger, F. Pfeiffer and H. Ulbrich, “Sensors and control concept of a biped robot,” IEEE Trans. Ind. Electron., vol. 51, no. 5, pp.972-980, Oct. 2004.

[2] Q. Huang and Y. Nakamura, “Sensory reflex control for humanoid walking,” IEEE Trans. Robotics, vol. 21, no. 5, pp. 977-984, Oct. 2005.

[3] Y. Guan, E. S. Neo, K. Yokoi and K. Tanie, “Stepping over obstacles with humanoid robots,” IEEE Trans. Robotics, vol. 22, no. 5, pp. 958-973, Oct. 2006.

[4] K. Harada, S. Kajita, F. Kanehiro, K. Fujiwara, K. Kaneko, K. Yokoi and H. Hirukawa, “Real-time planning of humanoid robot’s gait for force-controlled manipulation,” IEEE/ASME Trans. Mechatron., vol. 12, no. 1, pp. 53-62, Feb., 2007.