Desarrollo de un sistema de visión artificial asociado con el robot ANNO RV624 para reconocimiento y clasificación de piezas

Abstract

The main objective of this project was to create an artificial vision system associated with the ANNO RV624 robot to use for the recognition and classification of parts, for which a dedicated software and intercommunication of the robot with the artificial vision system was developed. This was done through the recognition of the objects through a web camera, which processed the image through an artificial vision algorithm allowing the recognition of the pieces and where they are located, after which the coordinates were sent to the controller and the robotic arm, thus obtaining that the robot organizes the pieces in an orderly manner. The starting point was the study of the operation, protocols, and actuators that the robot has, based on these criteria to establish possible ways of relating control systems and artificial vision. In the selected software platform, the algorithm for the evaluation of images that allowed the recognition of parts was codified; to this, control algorithms were associated for the linkage of these to the robotic actuator and to execute the classification of parts in the physical environment. An interface was generated where the user can easily establish the configurations and preferences to determine the functionality of the robot. The results achieved from this implementation is that the system worked autonomously, without having operators nearby to classify the parts. It is concluded that there is the possibility of making autonomous an industrial process of parts classification through a webcam with the help of artificial vision which improves the process times. It is recommended to work first with the original software of the robot to then create the software of own development, reviewing the manuals to use the libraries.