Many artificial intelligence applications need huge quantities of image and sensor data. Generating these is time-consuming. Image data that is generated artificially while remaining sensor-realistic reduces these efforts significantly and greatly speeds up the development of AI-based applications. Synthetic images, in particular, can be used to consider underrepresented cases and thus deliberately generate more balanced datasets.
To obtain these highly realistic images, GAN architectures show particularly promising initial results. However, their suitability must be evaluated using fitting metrics.
Determining optimal poses for measuring workpieces in 3D is a challenging and laborious task that often yields suboptimal results. By using reinforcement learning to plan 3D sensor measurements with the aid of models, optimal measurement poses for 3D measurements can be identified using CAD models of a wide variety of workpieces without having to perform real measurements.
Numerous CAD workpieces are required for training in order to derive effective measurement poses for a wide variety of component geometries. In addition to different 3D sensors, the optical features of the components are also taken into account in order to obtain complete and high-quality measurement data. Comprehensive measurement planning can be used to derive optimized strategies, e.g. in terms of component coverage.