Automated Quality Control of 3D Printed Tensile Specimen via Computer Vision

This research explores the integration of a robotic arm using computer vision for automated quality control for sorting 3D printed tensile specimens. The study, conducted, focuses on utilizing a Niryo NED-2 robotic arm with a vision system. The robotic arm captures cross-sections of tensile specimen, and a Python program processes vision feeds, filtering images based on 2D contours. Tensile samples were manufactured using Fused Deposition Modeling (FDM) with PLA material, incorporating known offsets (both positive and negative). Their dimensions were predicted and compared with the actual geometrical measurements. Experimental results showcase the system's accuracy in measuring specimen dimensions, demonstrating low error rates. The study highlights the potential for automated quality control in additive manufacturing, presenting a valuable tool for Industry 4.0. The robotic arm's vision system proves effective in enhancing efficiency and reliability in 3D printing quality inspection processes.