Integrated Robotic Arm Control: Inverse Kinematics, Trajectory Planning, and Performance Evaluation for Automated Welding

Abstract

This research delves into the automated functionality of robotic arm manipulators, a hallmark of Industry 4.0, within the manufacturing sector. The study focuses on precise movement adhering to predetermined trajectories, addressing the vital aspects of inverse kinematics and trajectory planning in robotic arm control. Utilizing the Matlab robotic toolbox, the research conducts simulations of inverse kinematic and trajectory planning. An experimental setup involving a robotic arm controlled by an Arduino Mega 2560 microcontroller, embedded with the inverse kinematic algorithm and trajectory planning, is executed. Data acquisition involves inputting coordinates and orientation for automatic welding along a straight path. Joint angles are measured using rotary encoders and converted into Cartesian coordinates to determine the end-effector's position. Discrepancy analysis compares measured joint angles with simulation values, revealing error margins. Movement quality of the robotic arm is assessed through Capability Processes (CP) evaluation. Results indicate disparities between experimental and simulated values. At input coordinates (400mm, 0mm, 300mm), joint angle errors of 2.51º, 0.98º, and 1.48º are observed for joints 2, 3, and 5, respectively. Similarly, at input coordinates (300mm, 0mm, 300mm), joint angle errors of 1.17º, 1.5º, and 2.7º are registered for the same joints. Trajectory error analysis during straight welding reveals average errors of 2.25 mm and 10.57 mm along the x and y axes. Mean absolute errors for joints 2, 3, and 5 are 1.9º, 0.48º, and 1.91º.

Keywords: Robotic Arm Manipulators, Inverse Kinematics, Trajectory Error Analysis