6-DOF Robot Arm Visual Servoing with Canny Edge-Based Object Detection
Dong Hyeon Kim, Min Cheol Lee · 2021 21st International Conference on Control, Automation and Systems (ICCAS) · 2021
Recently, the robot industry has developed and is doing human work instead. Among them, Visual Servoing research is actively underway to control robots using vision sensors. In this study, using OpenCV, a machine vision open-source library, to detect the contours of the object and the central point of the object. Machine Vision algorithms are performed using features such as Grayscale, Gaussian filter, Canny Edge, Contouring. The proposed algorithm recognizes only objects on the workstation without detecting other things. Also, transform the coordinate system of the vision sensor into the robot arm's coordinate system so that the robot arm can move correctly around the center of the object. For this purpose, camera calibration was performed, and a transformation matrix about vision to robot was obtained. Furthermore, we aim to control the 6-DOF robot arm through experiments to move the robot arm's end-effector to the center point of the object detected by the proposed algorithm.