Vision System for a Kuka KR-5 Industrial Manipulator

Abstract

In this article, we present the design of a vision system to take objects without previous knowledge. It is considered that these objects do not have a certain position. The manipulation will be carried out by an industrial robot Kuka KR 5, which does not have a vision system, so a Microsoft Kinect camera is used, as well as the PCA (Principal Component Analysis) algorithm to estimate the position and orientation of the objects within the robot's workspace. Among the tasks to be developed is the communication between the camera and the manipulator, the integration of additional components such as; ROS (Robot Operating System), Open CV, Solidworks and MatLab-SimMechanics. In the first stage, the design of the system as whole is presented. In a following work, the manipulative task (pick and place) is carried out with the manipulator robot.