ABSTRACT:

The development of vision-based object sorting manipulators has significant attention due to their potential applications in various industrial and domestic environments. The study presents a comprehensive framework for a vision-based robotic manipulator designed for sorting objects based on their visual characteristics. The system integrates advanced image processing techniques and machine learning algorithms to identify and classify objects in real- time. Key components include a high-resolution camera, a robotic arm equipped with multi- degree-of-freedom manipulators, and a central processing unit for data analysis and control. Once an object is classified, the robotic arm, guided by precise control algorithms, sorts the objects into designated categories. The vision-based sorting manipulator shows promise for enhancing automation in manufacturing, recycling, and inventory management, providing a robust solution for efficient and intelligent object handling. Through the project, students engaged in a cross-functional team approach via the 3D EXPERIENCE platform, gaining hands-on experience with DELMIA software and robotic simulation, as well as developing an image classifier AI model. These practical experiences underscore the project’s educational and technological contributions to the field of automated object sorting.