Vision servo based Delta robot to pick-and-place moving parts

Abstract

A vision-servo system is developed to make a Delta robot pick and place parts on a moving conveyer. To achieve motion planning of the Delta robot under a proposed vision servo system, forward and backward kinematics of the Delta robot is derived at first. Then, the vision servo program is developed in C++ to perform the image processing and the image recognition. For the image processing procedures, Canny and Sobel edge detection methods are studied. Based on the experimental results, Canny edge detection is better than the Sobel method in this case. Finally, image processing, image registration and motion controller movements are integrated into one system to perform the vision-servo pick-and-place the parts in the conveyor automatically.