Automatic Disassembly System using Computer Vision for the Omron Viper 650 Robotic arm

This paper presents an automated disassembly system using the Omron Viper 650 robotic arm, focused on separating two plates joined by four screws. In the first phase, different end effectors were designed and built using SolidWorks and 3D printing technology, addressing the technical challenges of accurately extracting the plates. In the second phase, a vision system was integrated, based on advanced computer vision and machine learning techniques, using Python and OpenCV. The integration of these systems allowed the robotic arm to autonomously detect, locate, and disassemble the plates, regardless of their initial orientation. The use of AI models, such as the YOLO algorithm, improved the accuracy and efficiency of the process. The results demonstrated that the combination of the mechanical design of the end effectors and artificial vision can effectively separate the assembly into its different parts, contributing to more efficient and sustainable disassembly processes.