Reactive Grasping using Optical Proximity Sensors

Abstract: We propose a system for improving grasping using optical proximity sensors mounted on fingertips that allows for online grasp adjustments to an initial grasp point without the need for premature object contact or regrasping strategies. Our system is composed of a novel optical proximity sensor for pre-touch pose estimation, a probabilistic model for robustly combining sensor readings, and a hierarchical reactive controller. This system can be used to complement existing grasp planning algorithms, or be used in more interactive settings where a human indicates the location of objects. We perform a series of experiments using a three-fingered hand equipped with our custom designed sensors to grasp a variety of common objects with mixed geometries and surface textures. The results validate the system as a means for improving the stability and quality of existing grasp strategies.