A statistical representation of three-dimensional shapes is introduced, based on a novel four-dimensional feature. The feature parameterizes the intrinsic geometrical relation of an oriented surface-point pair. The set of all such features represents both local and global characteristics of the surface. We compress this set into a histogram. A database of histograms, one per object, is sampled in a training phase. During recognition, sensed surface data, as may be acquired by stereo vision, a laser range-scanner, etc., are processed and compared to the stored histograms. We evaluate the match quality by six different criteria that are commonly used in statistical settings. Experiments with artificial data containing varying levels of noise and occlusion of the objects show that Kullback-Leibler and likelihood matching yield robust recognition rates. The present study proposes histograms of the geometric relation between two oriented surface points (surflets) as a compact yet distinctive representation of arbitrary three-dimensional shapes.

No abstract

We present a method for transferring grasps between objects of the same functional category. This transfer is intended to preserve the functionality of a grasp constructed for one of the objects, thus enabling the analogous action to be performed on a novel object for which no grasp has been specified. Manipulation knowledge is hence generalized from a single example to a class of objects with a significant amount of shape variability. The transfer is achieved through warping the surface geometry of the source object onto the target object, and along with it the contact points of a grasp. The warped contacts are locally replanned, if necessary, to ensure grasp stability, and a suitable grasp pose is computed. We present extensive results of experiments with a database of four-finger grasps, designed to systematically cover variations on grasping the mugs of the Princeton Shape Benchmark.

Abstract-Multi-fingered robot grasping is a challenging problem that is difficult to tackle using hand-coded programs. In this paper we present an imitation learning approach for learning and generalizing grasping skills based on human demonstrations. To this end, we split the task of synthesizing a grasping motion into three parts: (1) learning efficient grasp representations from human demonstrations, (2) warping contact points onto new objects, and (3) optimizing and executing the reach-and-grasp movements. We learn low-dimensional latent grasp spaces for different grasp types, which form the basis for a novel extension to dynamic motor primitives. These latent-space dynamic motor primitives are used to synthesize entire reach-and-grasp movements. We evaluated our method on a real humanoid robot. The results of the experiment demonstrate the robustness and versatility of our approach.

No abstract