Masters Theses

Date of Award

5-2022

Degree Type

Thesis

Degree Name

Master of Science

Major

Mechanical Engineering

Major Professor

William R. Hamel

Committee Members

William R. Hamel, Bradley Jared, Caleb Rucker

Abstract

A typical wire arc additive manufacturing (WAAM) robot cell consists of a 6 DOF robot manipulator and a 2 DOF part positioner. Since the WAAM process requires a minimum of 5 DOFs, there are three redundant DOFs in the system that can be utilized to improve the robot manipulator positioning during part printing. In this thesis, the redundant kinematics of a manipulator and part positioner robot system are solved and then implemented on an actual robot system. The inverse kinematics of the manipulator and part positioner are solved as a kinematic chain using the pseudo-inverse Jacobian method. The two DOFs in the part positioner are solved using a gravity aligned constraint function, which ensures that the welding torch is always vertically aligned and welding normal to the part path. The last DOF, the arbitrary z-rotation of the weld torch, is solved using the gradient projection method, which projects a cost function onto the nullspace of the Jacobian. A singularity avoidance cost function and a joint limit avoidance cost function were tested and compared. Additionally, the kinematics where solved with the z-rotation of the weld torch set at a fixed angle relative to the world base, effectively minimizing manipulator movement during the path. Comparing the solved joint angles shows that the singularity avoidance cost function struggled to create a smooth and collision free part path. The fixed torch rotation joint solutions gave the best robot position results, which were then tested on the actual robots and compared to the solutions found by a separate robot programming software.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS