Segmentation and Manipulation of Cork Strips in Bulk

Abstract

The production of cork stoppers is the largest application of natural cork, which is an ever-growing industry sector. Many attempts have been made to increase the automation of this process, such as the use of automated cork punching machines, but not all steps of this process are fully efficient such as the manipulation of cork strips prior to perforation, which is still a hand labor. This paper presents a system based on an RGBD camera and a 6 DoF robotic arm that manipulates cork strips which are disposed in bulk, either in a container or in a conveyor belt. It uses computer vision techniques to segment a single cork strip from the bunch and motion planners to control the robotic arm in order to grab the selected cork strip. On the experiments made, the system was able to correctly grab a cork strip with 92% success rate and with a frequency of 6 strips per minute.