6 DOF 3D PRINTED ROBOT FOR VISION BASED SORTING

Abstract

This paper presents a low-cost and efficient object sorting system using a fully 3D-printed robotic arm with six degrees of freedom (6DOF). The system uses two types of sorting methods: sensor-based sorting and camera-based (vision-based) sorting. In the sensor-based method, sensors and servo-controlled deflectors on the conveyor belt help in allowing or blocking objects based on their physical properties like shape and color. In the vision-based method, a camera controlled by a servo captures images of the objects, and a Convolutional Neural Network (CNN) processes these images to identify the object’s shape, color, and type. Only the correctly identified objects are sent to the robotic arm for picking and placing. The 3D-printed arm is cost-effective and offers flexibility in design. Tests show that the system works accurately and quickly in different conditions. This sorting system is useful for industrial automation, education, and research. The integration of both vision and sensor-based sorting increases reliability and reduces error rates. The modular nature of the system allows for easy upgrades and maintenance. The robotic arm’s motion is precisely controlled based on real-time object data, improving the overall efficiency of sorting tasks. The use of 3D printing also allows quick customization for specific applications. This system demonstrates a strong potential for smart and automated manufacturing environments.