Multi-point static dexterous posture manipulation for the stiffness identification of serial kinematic end-effectors.
Loading...
Date
2020
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The low stiffness inherent in serial robots hinders its application to perform advanced operations due to its reduced accuracy imparted through deformations within the links and joints. The high repeatability, extended workspace, and speed of serial manipulators make them appealing to perform precision operations as opposed to its alternative, the CNC machine. However, due to the serial arrangement of the linkages of the system, they lack the accuracy to meet present-day demands.
To address the low stiffness problem, this research provided a low-cost dexterous posture identification method. The study investigated the joint stiffness of a Fanuc M10-iA 6 Degree of Freedom (DOF) serial manipulator. The investigation involved a multivariable analysis that focused on the robot’s workspace, kinematic singularity, and dexterity to locate high stiffness areas and postures. The joint stiffness modelling applied the Virtual Joint Method (VJM), which replaced the complicated mechanical robot joints with one-dimensional (1-D) springs.
The effects of stress and deflection are linearly related; the highest stress in a robot’s structure is distributed to the higher load-bearing elements such as the robot joints, end-effector, and tool. Therefore, by locating optimal postures, the induced stresses can be better regulated throughout the robot’s structure, thereby reducing resonant vibrations of the system and improving process accuracy and repeatability. These aspects are quantifiably pitched in terms of the magnitude differences in the end-effector deflection.
The unique combination of the dexterity and the stiffness analyses aimed to provide roboticists and manufacturers with an easy and systematic solution to improve the stiffness, accuracy, and repeatability of their serial robots. A simple, user-friendly and cost-effective alternative to deflection measurements using accelerometers is provided, which offers an alternative to laser tracking devices that are commonly used for studies of this nature.
The first investigation focused on identifying the overall workspace of the Fanuc M-10iA robot. The reachable workspace was investigated to understand the functionality and potential of the Fanuc robot. Most robotic studies stem from analysing the workspace since the workspace is a governing factor of the manipulator and end-effector placement, and its operations, in a manufacturing setting.
The second investigation looked at identifying non-reachable areas and points surrounding the robot. This analysis, along with the workspace examination, provided a conclusive testing platform to test the dexterity and stiffness methodologies. Although the research focused on fixing the end-effector at a point (static case), the testing platform was structured precisely to cater for all robotic manufacturing tasks that are subjected to high applied forces and vibrations. Such tasks include, but are not limited to, drilling, tapping, fastening, or welding, and some dynamic and hybrid manufacturing operations.
The third investigation was the application of a dexterous study that applied an Inverse Kinematic (IK) method to localise multiple robot configurations about a user-defined point in space. This process was necessary since the study is based on a multi-point dexterous posture identification technique to improve the stiffness of Serial Kinematic Machines (SKMs). The stiffness at various points and configurations were tested, which provided a series of stiff and non-stiff areas and postures within the robot’s workspace.
MATLAB®, a technical computing software, was used to model the workspace and singularity of the robot. The dexterity and stiffness analyses were numerically evaluated using Wolfram Mathematica.
The multivariable analyses served to improve the accuracy of serial robots and promote their functionality towards high force application manufacturing tasks. Apart from the improved stiffness performance offered, the future benefit of the method could advance the longevity of the robot as well as minimise the regular robot maintenance that is often required due to excessive loading, stress, and strain on the robot motors, joints, and links.
Description
Masters Degree. University of KwaZulu-Natal, Durban.