An Artificial Neural Network-based Approach for Inverse Kinematics of PUMA 260 Robot
Keywords:
Artificial Neural Network, Denavit Hartenberg, End-Effector Cartesian position, Inverse Kinematics, PUMA 260 Robot Manipulator, Robotics Vision ControlAbstract
This study addressed the use of Artificial Intelligence (AI) techniques to solve the Inverse Kinematics (IK) and analysis issues for robotic manipulators, particularly PUMA 260 models with 6-Degrees of Freedom (6-DOF). A Robotics Vision Control (RVC) toolbox simulation in MATLAB was utilised to evaluate the Denavit Hartenberg (DH) and Forward Kinematics (FK) of the robot. Furthermore, the relationship between the joint angles of rotation and the end-effector Cartesian positions of the robotic manipulator was investigated. In reducing the complexity of the IK analysis, an Artificial Neural Network (ANN) was applied to estimate the IK of the robot using the Neural Network (NNtool) toolbox in MATLAB. This study successfully demonstrated the efficiency and applicability of the proposed ANN method for controlling robot motion by predicting IK parameters with high precision and minimal error. In determining the values of some IK joints, the results of the proposed technique significantly decreased to 1.579%. Thus, integrating the RVC and NNtool programmes considerably improved the accuracy in estimating ANN joint angles as a response to the input end effector positions. Consequently, the suggested ANN controller design was simpler, inexpensive, and more accurate in estimating the joint angles of the robot than standard regulating approaches. Based on the results, this study effectively and practically addressed the IK issues in robotic manipulators.
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2024 International Journal of Integrated Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Open access licenses
Open Access is by licensing the content with a Creative Commons (CC) license.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.