Development of a framework for an automated mechanical testing

• Autor(a) principal: Nunes, Diego Saraiva
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/31331
• topic: Specimen handling; Automation; Material testing; LabVIEW; Robot control; Calibration; ROS; ChArUco

Nowadays, predict or characterize material's behaviour are arduous tasks which involve performing repetitive tests using different testing machines and sensors (internal, such as load cells, and externals, such as cameras that measure deformations). When performed manually, they result in inaccurate data, due to the repeatability of the process. However, these handling specimen tasks could be performed automatically using traditional robotic manipulators, and creating systems based on automatic mechanical testing. Consequently, the precision of the testing results increases, as well as the automatism of the processes and the test's throughput. Some commercial products are already available on the market, but these options are offered in complete systems and cannot adapt themselves to existing equipment or retrofit systems. In this work, it is created a communication infrastructure between the equipment to be included in the automatic system designed. The whole process is managing the interface communication between devices and, consequently, composing the automatic material testing routine. This requires implementing several changes in the testing machine selected aiming to increase its degree of automation and, consequently, allow further integration in the fully automatic testing procedure. Additionally, a visual perception system is created using a specimen tray randomly positioned on a setup, which includes a camera and robotic manipulator that is automatically operating the traditional specimen handling of testing. The calibration methods are accomplished using ROS framework and the final system achieved operates in automatic mode, handling specimens from the tray prototype (designed for this work) to a fixed position previously taught to the manipulator, which represents the exact feeding position in machine tests.