The Realization of a Collaborative Robot ACP System for Intelligent Garbage Classification

This paper addresses the global waste management crisis by proposing an innovative strategy that combines visual recognition technology with robotics, grounded in the ACP methodology, aiming to create an intelligent garbage classification system. The primary goal of the system is to handle the growing volume of waste by employing the YOLOv7 object detection model in conjunction with a Sawyer/Baxter collaborative robot to achieve effective garbage categorization. The study simulates the application process of ACP in a laboratory setting. Robots can select different 3D-printed soft grippers based on needs. Collaborative robots generate new data through actual operations, which is then used to update and improve the model, forming a continuous iterative closed-loop system. Despite environmental challenges, the system’s achievements indicate potential for large-scale applications. Future research will focus on refining detection models, enhancing human-robot collaboration, and ensuring scalability for broader integration into waste management systems.