Neural Network Model for Identifying Workspace, Forward and Inverse Kinematics of the 7-DOF YuMi 14000 ABB Collaborative Robot

This research investigates development of a visual and analytical tool for the study of the seven degree of freedom (7-DOF) ABB 14000 YuMi robot. It provides an understanding of the manipulator workspace regions where they are visually represented for insight into the true work window. The developed tool is capable of realizing forward and inverse kinematics as well. The task is achieved using artificial neural networks that predict an inverse kinematic solution within an acceptable confidence interval (90th-95th percentile). Mathematical models, including all kinematic models i.e. the physical structure of manipulator joints and links, are developed and visually represented in the MATLAB platform by using the neural network toolbox. (C) 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.