Evaluation of coding and classification systems in the design of robotic grippers
Date
Authors
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Abstract
This study has been conducted to take advantage of geometrical similarity of parts within a family in the design of a robotic gripper set for the quick-gripper-changer. The group technology concept was used to define the part families. Four different coding and classification methods are used to analyze and define the sets of grippers. These methods include two different coding methods and two different classification methods. The coding methods include production flow analysis (PFA) based on a process routing and the Opitz system based on a part geometry. The classification methods include rank order cluster analysis (ROCA) and cluster analysis with similarity coefficients (CASA). The following four methods have been used in this study to group parts into families. (1) Production flow analysis/Rank order cluster analysis. (2) Production flow analysis/Cluster analysis with similarity coefficient. (3) Opitz/Rank order cluster analysis. (4) Opitz/Cluster analysis with similarity coefficient;Softwares for the four coding and classification methods were developed in this research. By means of the software, part families were defined. A gripper was configured for each part family per method. The gripper features included in this study were the gripping mechanism and the jaw shape. The gripper types were two finger mechanical gripper, a magnetic gripper, and a vacuum gripper. Each part within part families was evaluated whether it could be grasped successfully by the gripper configured for the part family. Based on the number of parts which were grasped successfully, the robotic gripper set was defined for each method. The reasonable percentage of parts which could be grasped by the robot gripper was obtained for each coding and classification method;The Opitz/ROCA method showed the highest percentage of parts which were successfully grasped by the defined robotic gripper set. The set included a twin "V" shape notched mechanical gripper, a twin plane shape mechanical gripper, and a vacuum gripper. The 74.3 percentage of parts were grasped successfully by the defined gripper set.