Diseño y simulación de un sistema automatizado basado en visión por computador para una estación de paletizado usando Robots colaborativos.

Abstract

In this project, an automated palletizing station is developed and simulated using RoboDK to optimize the recognition, classification, data storage and handling of products previously cataloged by QR codes. This automated solution seeks to improve efficiency and accuracy in the palletizing process, responding to the industrial need to organize, register and prepare products for storage. The implementation of this technology is considered essential to reduce processing time and minimize human error, which is crucial in a highly competitive industrial environment. For the development of the project, two cameras are used for image acquisition, these images are processed through Python programming. Using the OpenCV library, the necessary calculations are made so that, with the first camera, the position and orientation of each box on the conveyor belt is detected, this information is sent to the first robot so that it can pick it up when it reaches a midpoint of the conveyor belt, this robot has a sensor for the identification of the object. Once the first robot picks up the box, it proceeds to approach the box to the second camera, which, with Python programming, is processed to decode the QR code that indicates the type of product that has arrived, to be subsequently palletized according to the type of product. The first robot rotates the box on the first four faces accessible to the camera. If the QR code is not found on any of these faces, the signal is sent to the second robot, which approaches to pick up the box and rotates in such a way as to allow the second camera to access the other two remaining faces. Once the QR code is found, the signal is sent to the first robot to pick it up again and place it on the conveyor belt, which carries the box to the end of the conveyor belt. In the meantime, the information is also sent to the third robot, in charge of picking up the box and placing it on the pallet corresponding to the type of product detected, according to the QR code. With the first camera the edges of the object are detected, with which the center of the object is found, and finally, considering the X axis as a reference, the orientation is calculated, with this orientation the corresponding calculations are made for the orientation of the robot's hand so that it can pick up the object correctly. IV Keywords: Computer Vision, Robots, RoboDK, Python, OpenCV,