光學視覺與機械手臂系統整合之校正方法介紹

A Systematic Literature Review on Eye-Hand Calibration

在現代的機械手臂應用當中，與光學視覺系統搭配已經成為主流的趨勢，透過視覺回授伺服控制來達到機械手臂智慧化與自動化的目的。然而視覺回授控制仍需藉由手眼校正的演算法將相機座標系及機械手臂座標系對齊，根據相機的配置有分為Eye in hand以及Eye to hand。此篇文章首先定義手眼校正的兩種配置與數學模型建立，接著依照兩種定義整理不同種類的手眼校正方法、應用場域及實際配置案例，讓讀者能夠根據不同配置以及場域進行參考。

In the modern application of robotic manipulators, collocation with an optical system has become mainstream. However, to achieve vision servo control, the alignment between the coordinate of the camera and the manipulator should be calibrated. According to the placement of the camera, the method of the eye-hand calibration can be divided into eye-in-hand and eye-to-hand. In this article, the mathematic models of these two methods are built and the literatures are arranged by the order of scenario, optimized algorithms and the placement of cameras.