Haptic identification of objects using a modular soft robotic gripper


This work presents a soft robotic gripper capable of robustly grasping and identifying objects based totally on inner state measurements. A rather compliant hand permits for intrinsic robustness to grasping uncertainty, but the specific configuration of the hand and object is not known, leaving undetermined if a hold close used to be successful in choosing up the right object. A soft finger used to be tailored and mixed to structure a three finger gripper that can without difficulty be connected to current robots, for example, to the wrist of the Baxter robot. Resistive bend sensors were delivered inside every finger to furnish a configuration estimate enough for distinguishing between a set of objects. With one records point from every finger, the object grasped by way of the gripper can be identified. A clustering algorithm to find the correspondence for every grasped object is introduced for both enveloping grasps and pinch grasps. This hand is a first step closer to sturdy proprioceptive gentle grasping.