視覺與機械手臂整合於產業之應用

Integration of the Machine Vision and the Robotic Manipulators for the Industrial Application

蔡林承緯(Cheng-Wei Tsai Lin)

指導教授 ： 李政道

國立虎尾科技大學/工程學院/自動化工程系碩士班/碩士(2017年)

摘要

近年由於全球製造業面臨缺工及少量多樣生產的挑戰，如何導入多種自動化與智慧化元素已成為各廠商當前重要課題，例如：機器視覺取代人眼以縮短測量時間與降低檢測失誤率、機械手臂代替工人以提高生產效率和解決缺工問題、甚至是整場的自動化與工業4.0環境建構。因此，本研究將整合視覺系統與機械手臂開發一自動化上下料系統。首先，進行多關節機械手臂的運動學推導，並成功開發一機械手臂控制器進行驗證。視覺檢測將以靜態尋物系統與動態追物兩種方式進行實驗。在靜態尋物系統實驗中，先針對視覺系統拍攝到的檢測物進行影像處理，透過檢測物的當前座標並與機械手臂座標同步，再透過串列通訊方式完成機械手臂自動抓取檢測物的功能；動態追物實驗以靜態尋物系統之實驗架構為基礎，將利用視覺系統檢測物件得到時間差、位置差，進而計算出輸送帶轉速以預測出檢測物移動位置後，再利用機械手臂進行動態追物抓取之動作。最後，機械手臂上下料功能再利用串列傳輸與數位輸入/輸出模組與CNC控制器進行交握以完成自動化上下料與加工之功能。

關鍵字

機器視覺 ； 機械手臂 ； 自動化上下料系統

並列摘要

Customized manufacturing is increasing years by years, how the manufactory imports the automatic equipment to the production line is the important topic. The measurement time and the error of inspection is decreased when the machine vision replace the human source.
Besides, the study completed the system of automatic loading and unloading for automatic production by manipulator and machine vision. First, the study derived and proved the Kinematics of manipulator, but the study used the commercially available manipulator to experiment.
The study of vision inspection is divided into the target of static seek and the dynamic target tracking object to experiment. For the target of static seek, the picture of machine vision to image of processing, and corrected the coordinate of machine vision and manipulator to the same origin by the material’s position from the picture of the machine vision. After the coordinate synchronizing, the machine vision and manipulator completed the function the material of gripping by serial transmission. The experimental of dynamic target tracking use the concept of the target of static seek to complete the coordinate synchronize between the manipulator and the machine vision. Then, the velocity of the conveyor will be determined by time interlude and moving distance from machine vision in the program by using Labview software. At the same time, the program used the velocity of the conveyor and the position of material from machine vision to predict the position of target. The manipulator is loading and unloading the material when the program is issue the command. Finally, the study completed the function of automatic loading and unloading by serial transmission and I/O signal for the equipment.

並列關鍵字

machine vision ； manipulator ； system of automatic loading and unloading

參考文獻

[10]能麟企業股份有限公司官網，PAC可程式自動控制器。

[2]陳冠宇、張永鵬、傅培耕，2003年，即時影像追蹤系統之設計與實現，Vol.31,No. 1pp21-29。

[39]Ming-Chin Chuang, Digital Image Processing-Image Segmentation.

[11] Ik Sang Shin, Sang-Hyun Nam, Hyun Geun Yu, Rodney G. Roberts, 2006,” Conveyor Visual Tracking using Robot Vision”, 2006 Florida Conference on Recent Advances in Robotics, FCRAR 2006.

[4] A. Lipton, H. Fujiyoshi and R. Patil, 1998, “Moving Target Classification and Tracking from Real-time Video ”, IEEE Workshop on Applications of Computer Vision. Princeton, NJ.