手眼協調控制是人類靈活的操控技巧之一，本研究將使用體感偵測器人類雙眼產生的深度與色彩資訊，藉由分析三維空間資訊以及電腦的模擬控制，結合所設計的拍擊機構，實現人類的手眼協調控制技巧。經由探討手腕部分之手眼協調控制系統，能為未來智慧型機器人之開發建立基礎。
本文選用乒乓球之拍擊重現人類手臂及手腕靈活的操控技巧，使用體感偵測器取得色彩資訊，經過影像處理分離目標物及背景，結合深度資訊，將目標球體位置轉換成在世界座標中的空間資訊。利用空氣動力學、碰撞力學以及基本的運動方程式分析球體受力狀態並預測球體運動軌跡，經過擬定的適應性拍擊策略計算出拍擊角度及修正角度，輸出控制指令至控制板控制機械手臂，使機械手臂完成指定拍擊動作，以達到模擬人類手臂連續拍擊之目的。
最後，利用所建構的拍擊系統，進行拍擊實驗，目的是為了測試拍擊機構及系統之即時性，以及拍擊策略能否改善實際拍擊所產生的振動問題，並測試系統自動修正的適應能力。由實驗結果可看出，本研究所使用的適應性拍擊系統具有足夠的即時性，能應付連續拍擊時球拍的角度快速變換，拍擊策略可降低因震動對拍擊造成的影響，而達到連續拍擊的目的。

Hand-eye coordination control is one of the dexterous operational skills of mankind. This study applies an RGB-D camera to sense human stereoscopic vision, including information of depth and color. Through analyzing three-dimensional spatial information with the help of computer simulation, and integrating with a robotic arm, the hand-eye coordination control for bouncing a ball using the robotic system is realized. The achievement of this study provides the foundation for future development of intelligent robots.
This thesis aims at bouncing a table tennis ball to reproduce skillful movement of human arm and wrist. The target ball and the background are first separated by using the RGB-D camera to acquire color information. The target ball’s position in the world coordinate frame can therefore be obtained by incorporating the depth information. Then apply aerodynamics, collision mechanics and equations of motion to analyze forces on the ball and predict its future motion trajectory. An adaptive strategy calculating the bouncing angle and the corrected angle of the paddle board is developed to control the robotic arm so that the designated objective of bouncing ball can be achieved to simulate intelligent movement of human beings.
At last, experiments of bouncing a ball are conducted by employing the adaptive ball bouncing system. A number of issues, including the real-time response of the experimental system, improvement of vibration problem caused by bouncing the ball applying the presented bouncing strategy, and adaptive capability with automatic adjustment, are investigated. Experimental results show that the proposed adaptive ball bouncing system displays sufficient real-time feature to cope with rapid angle changes of the paddle board during continuous bouncing. Besides, the adaptive bouncing strategy reduces vibration and achieves the purpose of continuous bouncing.

目錄i
圖目錄iv
表目錄vi
摘要vii
Abstractviii

第一章 緒論1
1.1 研究動機與目的1
1.2 研究方法與步驟1
1.3 文獻回顧4
1.4 論文結構6
第二章 硬體設備與架構7
2.1 攝影鏡頭與機械手臂架設7
2.2 ASUS Xtion Pro LIVE10
2.3 RGB-D Camera 原理11
2.4 控制板型號12
2.5 機器手臂與馬達規格13
2.6 拍擊板15
2.7 乒乓球17
第三章 影像處理演算法與深度資訊18
3.1 RGB色彩空間與灰階影像18
3.2 HSV色彩空間19
3.3 影像二值化20
3.4 侵蝕(Erosion)與膨脹(Dilation) 22
3.5 開放運算與封閉運算23
3.6 深度資訊24
3.7 影像處理流程25
第四章 力學分析與擊球策略27
4.1 球體受力情形27
4.1.1 空氣浮力28
4.1.2 空氣阻力28
4.1.3 碰撞力學30
4.1.4 恢復係數31
4.1.5 球體運動狀態模擬33
4.2 拍擊動作34
4.3 拍擊角度計算36
4.3.1 垂直拍擊36
4.3.2 向中心拍擊38
4.3.3 加入隨機變數模擬向中心拍擊40
4.4 適應性拍擊機制43
4.4.1 角度修正43
4.4.2 拍擊速度修正44
4.4.3 電腦模擬拍擊45
4.5 綜合拍擊實驗47
第五章 結論與未來規劃53
5.1 結論53
5.2 未來規劃53
參考文獻55

[1] 蘇恕德，兩段式背景相減法偵測移動物體，國立台灣大學電機資訊學院電機工程學研究所碩士論文，民國九十八年六月。
[2] 賴丙全，利用多攝影機進行移動物三維定位及追蹤，國立中央大學土木工程研究所碩士論文，民國九十六年七月。
[3] 蔡承訓，二軸機器手腕之影像伺服控制，國立中山大學機械與機電工程學系研究所碩士論文，民國九十六年七月。
[4] 周景荃，球板系統之影像伺服控制及路徑規劃，國立中山大學機械與機電工程學系研究所碩士論文，民國九十八年七月。
[5] T. Kizaki and A. Namiki, “Two ball juggling with high-speed hand-arm and high-speed vision system,” 2012 IEEE International Conference on Robotics and Automation, Saint Paul, MN, USA, 14-18 May, 2012, pp. 1372–1377.
[6] H. H. Rapp, “A ping-pong ball catching and juggling robot: A real-time framework for vision guided acting of an industrial robot arm,” 2011 IEEE International Conference on Automation, Robotics and Applications, Wellington, New Zealand, 6-8 December, 2011, pp. 430–435.
[7] P. Yang, Z. Zhang, H. Wang and D. Xu, “Design and motion control of a ping pong robot,” the 8th World Congress on Intelligent Control and Automation, Jinan, China, 7-9 July, 2010, pp.102–107.
[8] Z. Zhang, D. Xu and J. Yu, “Research and latest development of ping-pong robot player,” the 7th World Congress on Intelligent Control and Automation, Chongqing, China, 25-27 June, 2008, pp. 4881–4886.
[9] O. Birbach, U. Frese and B. Bauml, “Realtime perception for catching a flying ball with a mobile humanoid,” 2011 IEEE International Conference on Robotics and Automation, Shanghai, China, 9-13 May, 2011, pp. 5955-5962.
[10] 劉安陞，球彈跳之智慧型手眼協調控制，國立中山大學機械與機電工程學系研究所碩士論文，民國一百年七月。
[11] L. Cruz, D. Lucio and L. Velho, “Kinect and RGBD images: Challenges and applications,” 2012 XXV SIBGRAPI Conference on Graphics, Patterns and Images Tutorials, Ouro Preto, Brazil, 22-25 August, 2012, pp. 36–49.
[12] 周建佑，基於L-K運算法及Kinect的動態目標追蹤系統之研究，國立中央大學電機工程學系碩士論文，民國一百零一年六月。
[13] I. Siradjuddin, L. Behera, T. M. McGinnity and S. Coleman, “A position based visual tracking system for a 7 DOF robot manipulator using a Kinect camera,” 2012 IEEE World Congress on Computational Intelligence, Brisbane, Australia, 10-15 June, 2012, pp. 1–7.
[14] B. R. Munson, D. F. Young, T. H. Okiishi and W. W. Huebsch, Fundamentals of Fluid Mechanics, 5th Edition, John Wiley& Sons, 2006.