A Novel Design of a Robotic Object Sorter Based on Color Differences using Image Processing Techniques

Abstract

In the industrial production systems, manufacturers often face difficulties in sorting out colored objects. Human eyes cannot continuously differentiate colors, as they get tired very soon. Using manual sorting, many laborers are needed which can be a huge expense. The aim of this project is to build a robotic sorter, which is capable of detecting and sorting colored objects. It has a robotic arm, which, after detection of color, picks up the object and then place it at a predetermined place as required by the production system. This project is a combination of electrical, mechanical and visual subsystems. A Pixy cam along with the image processing software do the job of color detection. We have used several servomotors for rotating the arms of the robot. The arms can move horizontally from its base and vertically up and down. The gripper is capable of moving quite far for proper gripping of objects. This multi-DOF (Degree of Freedom) robotic sorter can be a very useful tool in fully automating a production process, which uses conveyer belts. The overall improvement in the efficiency of the production process can be significant by using this machine.