Dual stage-controlled robotic system for precise inline 3D measurements on moving objects

Abstract

To accomplish the demands for increasing product quality, an efficient 100% quality control for products with structures on the single- or even sub-micrometre range is desired. Robotic inline 3D measurement systems play a particularly important role in this context, as they can enable the flexible sample inspection at arbitrary measurement spots. However, as the positioning precision of industrial robots as well as environmental disturbances cause relative motion between the measurement tool on the industrial robot and the sample in the range of several tens of micrometres, the integration of robotic 3D measurement systems with sub-micrometre precision into an industrial production line is considered as a major challenge. Therefore, a robotic 3D measurement system including an electromagnetically levitated sample tracking measurement platform to maintain constant position of the integrated 3D measurement tool relative to a sample has been proposed recently. In this thesis, the provided robotic 3D measurement system is advanced towards precision measurements on moving objects by means of a dual stage-control concept, such as required for future inline applications. The relative position between the sample tracking measurement platform and the industrial robot measured by the internal position sensor system is used to precisely reposition the industrial robot, maintaining the measurement platform within its actuation range. A sophisticated control architecture combining feedback control as well as the conveyor velocity as a priori knowledge in a feed forward approach is designed and evaluated. The performance of the dual stage-controlled robotic system is identified by tracking a moving sample on a conveyor system. Residual sample tracking errors of 486nm in motion and 167nm in the vertical direction at a sample velocity of 10 mm/s are achieved. With the moving sample being actively tracked, the system is capable of performing 3D measurements with resolutions down to 620nm, achieving submicrometre precision.