| Resumo: | The programming of automation systems, namely robotic systems, is generally too complex and time spending for the industrys current needs. Particularly, in the most developed countries, where the industry is driven by small production series. To this kind of factory model the setup time is far too high compared to the actual production time. Therefore it is necessary to develop new techniques and make use of high end sensors that allow very flexible solutions. In this paper, it will be presented a low-cost system that obtains the 3D model of an object through laser/camera triangulation and then calculates the objects position and orientation by means of a non-analytical cost function, which is minimized through conjugate-gradient algorithm. The position result can then be send to a robotic manipulator. Concepts like camera calibration, image processing, 3D reconstruction and data merge will be presented in this article.
|
|---|