視覺控制機械手臂隨機取放系統之發展

Development of a Random Bin Picking Control System for a Robotic Arm Using Machine Vision

張均壕(Jun-Hao Zhang)

指導教授 ： 陳冠宇

中原大學/工學院/機械工程研究所/碩士(2018年)

https://doi.org/10.6840/cycu201800703

摘要

近幾年，工業4.0帶起新一波的智慧製造浪潮，其中具備視覺功能的機械手臂扮演重要角色，因此本文提出結合機器視覺發展具有形狀與特定顏色辨識的機械手臂隨機取放系統，以提升六軸工業機械手臂的應用功能。本文架設一具網路攝影機拍攝工作區域內的影像，經由色彩空間轉換、灰階化、二值化、影像平滑、霍夫轉換…等影像處理方法，進行物體的形狀偵測、顏色辨識與位置計算，所得結果經由機器人逆向運動學轉換成機械手臂各軸的旋轉角度，使機械手臂可以完成物體隨機取放的功能。本文選定三種特定顏色配合三種不同半徑之圓柱體工件進行實驗，在無加裝輔助光源的環境下，實驗成功率可達90%。

關鍵字

機器視覺 ； 形狀偵測 ； 顏色辨識 ； 數位影像處理 ； 逆向運動學

並列摘要

In recent years, Industry 4.0 has brought a new wave of smart manufacturing, in which the robotic arm with visual function plays an important role. Therefore, this dissertation proposes to develop a random bin picking control system with shape and specific color recognition based on machine vision for enhancing a wide range of applications of a six-axis industrial robotic arm. First, a webcam is set up to capture images in the working area in the dissertation. Then, shape detection, color recognition and position calculation of objects are performed by image processing methods such as color space transformation, grayscale conversion, binarization, image smoothing, and Hough transform. Finally, the obtained result is converted into the rotation angle of each axis of the robotic arm by the inverse kinematics, so that the random bin picking control system for the robotic can perform well. In this dissertation, three kinds of cylinders with different radii and different colors are used in experiments. Experimental results show that the success rate can reach 90% in the environment without auxiliary light source.

並列關鍵字

machine vision ； shape detection ； color recognition ； digital image processing ； inverse kinematics

參考文獻

參考文獻

[1] D. A., Forsyth and J. Ponce, Computer Vision: A Modern Approach, Pearson Publishing: London, U.K., 2002.

[2] 江修, 黃偉峰, “六軸機械手臂之控制理論分析與應用,” 智慧機械技術專輯, 2006.

[3] J. Carnicelli, “Stereo vision: Measuring object distance using pixel offset,” http://jvcai.blogspot.com/2005/08/stereo-vision-measuring-object-distance.html, 2005.

[4] A. Rahman, A. Salam, M. Islam, and P. Sarker, “An image based approach to compute object distance,” International Journal of Computational Intelligence Systems, Vol. 1, No. 4, pp. 304-312, 2008.