機器人技術變得更加先進且持續解決一些人類生活中的問題，像是安全、福利和醫療照顧。為了解決上述的問題，有好幾個人形機器人正被相繼地研發。本研究中，利用本實驗室過去所建構的一個輪式人形機器人，當作一個視覺導引抓取任務的研究平台。機器人是由一個輪式移動平台、架於輪式移動平台上的一個三自由度之身軀、兩隻七自由度之機械手臂、一個七自由度之機械右手掌、一個雙自由度之機械左手掌和一個七自由度之雙眼機械頭所構成。在雙眼機械頭中，每兩個攝影機各自構成一個完整的組合眼，其中上方的攝影機為一個擁有寬視野的筆形攝影機，而下方為一般的窄視野攝影機。因此，本研究選定機械頭上的三個攝影機作為一個共平面的類複眼。本論文的目標是提出一個視覺導引控制策略，以賦予人形機器人利用類複眼驅動手臂夾取物體的能力。所提出的控制策略是基於三攝影機的幾何模型與其相對位置，以及最小平方的技術。最後，利用一個物件夾取的實驗，驗證由機器人的類複眼進行視覺導引任務之效果。

The robot technology is becoming more advanced and is continuing the challenge towards solution of several problems of mankind, such as security, welfare, and medical care. Several humanoid robots are being developed successively to solve the problems. In this study, a wheeled humanoid robot that was constructed in our laboratory is used as a research platform for vision-guided grasping tasks. The robot consists of a wheeled mobile base, a torso with a 3 degrees of freedom (DOF) waist mounted on the mobile base, two 7 DOF robot arms, one 7 DOF robotic right hand, one 2 DOF robotic left hand and a 7 DOF robotic binocular head. In the robotic binocular head, each fully assembled eye comprises two color CCD cameras, the upper lipstick camera containing a wide-angle lens and the lower camera containing a narrow-field foveal lens. Therefore, three cameras of the robotic head are arranged to be a coplanar compound-like eye. The objective of this thesis is to propose a vision-guided control strategy to equip the humanoid robot with the ability to grasp the material by the robot arm using the compound-like eye of the robotic head. The proposed control strategy is based on geometric camera models and the relative location of three cameras, and the least squares technique. Finally, one experiment is conducted to verify the effectiveness of vision-guided grasping task by a humanoid robot with a compound-like eye.

[1] A. Abbott, Dynamic Integration of Depth Cues for Surface Reconstruction from Stereo Images, PhD Thesis, Department of Electrical Engineering, University of Illinois at Urban-Champaign, USA, 1990.
[2] N. Ayache and F. Lustman, “Trinocular Stereo Vision for Robotics,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.13, No. 1, pp. 73-85, Jan. 1991.
[3] G. Flandin, F. Chaumette and E. Marchand, “Eye-in-hand/Eye-to-hand Cooperation for Visual Servoing,” Proceedings of the 2000 IEEE International Conference on Robotics and Automation San Francisco, Vol. 3, pp. 2741-2746, CA, Apr. 2000.
[4] E. Gurewitz, I. Dinstein and B. Sarusi, “More on the Benefit of a Third Eye for Machine Stereo Perception,” Proceedings of the 1986 IEEE International Conference on Pattern Recognition, Paris, France, pp. 512-515, Oct. 1986.
[5] Y. Kawai, et al., “Stereo Camera System-PataPata,” Proceedings of the Symposium on Image Recognition and Understading, Vol. 2, pp. 127-134, Japan, 1994 (in Japanese).
[6] C. H. Lai, Development of a Humanoid Robot and Its Self-Learning Control for Robotic Table Tennis, PhD Thesis, Department of Mechanical Engineering, Nation Cheng Kung University, ROC, 2010.
[7] Y. M. Liu, Construction and Intelligent Control of a Feeding Humanoid Robot and Its Application to Two-Arm Coordination for Material Handling, MS Thesis, Department of Mechanical Enginerring, National Cheng Kung University, ROC, 2010.
[8] R. P. Paul, Robot Manipulator: Mathematics, Programming and Control, The MIT Press, Cambrzdge, MA, 1981.
[9] Y. Sumi, Y. kawai, T. Yoshimi and F. Tomita, “3D Objective Recognition in Cluttered Environments by Segment-Based Stereo Vision,” International Journal of Computer Vision, Vol. 46, No.1, pp. 5-23, 2002.
[10] F. Tomita, et al., “R&D of Versatile 3D Vision System VVV,” Proceedings of the 1998 IEEE International Conference on System, Man, and Cybernetics, San Diego, CA, Vol. 5, pp. 4510-4516, Oct. 1998.
[11] R. D. Umesh and J. K. Aggarwal, “Binocular versus Trinocular Stereo,” Procceedings of the 1990 IEEE Internation Conference on Robotics and Automation, Cincinnati, OH, pp. 2045-2050, May 1990.
[12] Z. Zhang, “A Flexible New Technique for Camera Calibration,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, pp. 1330-1334, Redmond, USA, Nov. 2000.