立體視覺伺服機器手臂之智慧追蹤控制的研究

A study on Intelligent Tracking Control of Robot Manipulator Based-on Visual Servoing with Stereo Vision

蔡博仰(Po-Yang Tsai)

指導教授 ： 陳冠宇 ； 許政行

中原大學/工學院/機械工程學系/碩士(2007年)

https://doi.org/10.6840/cycu200700594

摘要

隨著科技的日新月異，視覺伺服技術的應用目前已廣泛的融入日常生活當中，然而僅以單眼視覺伺服卻有著無法量測物體之三維資訊之缺點，因此本文使用立體視覺伺服，透過二部攝影機所擷取之影像視野重壘區域重建物體之三維資訊，搭配四軸機器手臂利用影像處理、立體視覺、座標轉換、機器人運動學…等方法達成針對目標物進行靜態夾取、動態追蹤以及目標路徑追蹤之功能開發。另外，為了克服機器手臂運作時所產生之慣性，本文以梯形速度曲線結合三次多項式內插法設計一套加減速控制機制，針對機器手臂運作情形進行點對點之速度曲線規劃，由實驗結果驗證此設計已有效的降低慣性對於機器手臂運作時之影響。
為了使立體視覺伺服系統更為強健，本文透過攝影機校正技術得知真實之攝影機模型，以攝影機模型之內部參數與外部參數更新立體視覺計算，對於因鏡頭扭曲而產生之影像扭曲現象進行誤差補償。

關鍵字

機器手臂控制 ； 立體視覺伺服 ； 路徑追蹤

並列摘要

The application of visual servoing technique is used widely in daily life with the developing of technology. However, single visual servoing system cannot survey the 3D information. Therefore the stereo visual servoing technique, using the overlapping region of the pictures which are captured by two cameras to make the 3D reconstruction of object is adopted in this dissertation. Also the experiment collocates the methods such as image processing, stereo vision, coordinate transform, manipulator kinematics with the 4-axis robot to accomplish the developing of grabbing in static, tracking in motion, and path-tracking. Furthermore, in order to overcome the inertia which accompany with the robot operation, the mechanism which contains acceleration and deceleration is designed based on trapezoid curve method and cubic interpolation. And then uses it for the speed command in the experiment, the result shows the design can reduce the effect of the inertia efficiently when robot operation.
For the image distortion caused by the lens of the camera, the real camera model is acquired by the camera calibration technique firstly, and then let the intrinsic parameters and extrinsic parameters replace the previous stereo imaging model. Using the way not only compensate the error but make the stereo visual servoing system more robust.

並列關鍵字

path tracking ； robot control ； stereo visual servoing

參考文獻

[1] A. Currie, The History of Robotics, http://cache.ucr.edu/~currie/roboadam.htm.

[6] 傅培耕，即時物體追蹤之立體視覺導引自走車，碩士論文，中原大學機械工程學系，2004。

[7] 張嘉麟，模糊邏輯控制可搭乘電梯之立體視覺伺服輪型機器人，碩士論文，中原大學機械工程學系，2006。

[3] Y. Shirai and H. Inoue , “Guiding a robot by visual feedback in assembling tasks,” Pattern Recognition, Vol. 5, pp.99-108, 1973.

[5] S. Hutchinson, G.D. Hager, and P.I. Corke,“A tutorial on visual servo control,” IEEE Trans. On Robotics and Automation, Vol. 12, No. 5, pp.651-669, 1996.