The paper reports on a methodology for complete automation of the process of scanning the surface of an unknown object through the utilization of optical range sensors. An incremental model representing object surface and workspace occupancy is used to define an optimization strategy for selecting the best scanning viewpoints and generate adaptive collision-free scanning trajectories. The optimization strategy proposed attempts to select the viewpoints that maximize the knowledge of the object taking into account the status of the current model and the constraints associated with the range sensor and positioning system. One of the novel aspects of this work is the fact that viewpoint selection and scanning path definition are performed automatically on the basis of a simple volumetric model in situations where potential collisions may arise. The ability to control the quality of acquired data and the reduction in global scanning trajectories when compared with manual scanning operations are other important aspects. The proposed approach is sufficiently general to be used in different scanning system configurations.
Read full abstract