AI-powered object sorting mechanical arm with easy replaceability

Abstract

This project presents the design and implementation of an AI-powered object sorting mechanical arm that integrates decision-making capabilities of AI to enhance flexibility and efficiency in industrial automation. The proposed system eliminates the need for manual reprogramming by employing a large language model (LLM), Gemini 2.0, for object recognition and decision-making. Equipped with a 4-DOF robotic arm, high-resolution cameras, and a gripper mechanism, the system is capable of identifying, classifying, and placing objects with high accuracy. The approach demonstrates improvements in adaptability, reduced setup time, and sustainable applicability across various sectors, aligning with the needs of modern industry and society.