Publicação
Automatic 3D parts laser marking system
| Resumo: | As the world’s population increases, the challenge of continuing to respond to their needs is also growing. In this way, the industrial world is once again tested, leading to the emergence of Industry 4.0. It is in this sense that the AdaptMark project emerges, launched as a challenge by the JPM company to the University of Aveiro, where the objective of this is the research and development of a robotic laser marking system of parts of different geometries and materials, endowed with high intelligence, autonomy of operation, flexibility and adaptability. The project can be divided into three subsystems: identification subsystem, laser manipulation and marking subsystem and quality verification subsystem, and this dissertation addresses the identification subsystem, which aims to correctly identify the parts that arrive randomly on a conveyor belt, as well as the correct estimation of the position of marking on them. Regarding the identification subsystem, it can be divided into two parts: one related to the identification of the parts and the other related to the estimation of their pose. In the part of the pieces identification, several 3D descriptors were tested and, in the end, it is possible to verify that all the pieces are correctly identified. Regarding the part about estimating the pose of the pieces, we ended up with the classification of the pieces about their shape and, in this dissertation, only rectangular pieces are addressed. |
|---|---|
| Autores principais: | Fernandes, Diogo Barbosa |
| Assunto: | Computer vision Point cloud Object recognition 3D descriptors Laser marking Robotics |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Aveiro |
| Idioma: | inglês |
| Origem: | RIA - Repositório Institucional da Universidade de Aveiro |
| Resumo: | As the world’s population increases, the challenge of continuing to respond to their needs is also growing. In this way, the industrial world is once again tested, leading to the emergence of Industry 4.0. It is in this sense that the AdaptMark project emerges, launched as a challenge by the JPM company to the University of Aveiro, where the objective of this is the research and development of a robotic laser marking system of parts of different geometries and materials, endowed with high intelligence, autonomy of operation, flexibility and adaptability. The project can be divided into three subsystems: identification subsystem, laser manipulation and marking subsystem and quality verification subsystem, and this dissertation addresses the identification subsystem, which aims to correctly identify the parts that arrive randomly on a conveyor belt, as well as the correct estimation of the position of marking on them. Regarding the identification subsystem, it can be divided into two parts: one related to the identification of the parts and the other related to the estimation of their pose. In the part of the pieces identification, several 3D descriptors were tested and, in the end, it is possible to verify that all the pieces are correctly identified. Regarding the part about estimating the pose of the pieces, we ended up with the classification of the pieces about their shape and, in this dissertation, only rectangular pieces are addressed. |
|---|