Platform portable anthropomorphic grasping with the bielefeld 20-DOF shadow and 9-DOF TUM hand

Abstract

Abstract-We present a strategy for grasping of real world objects with two anthropomorphic hands, the three-fingered 9-DOF hydraulic TUM and the very dextrous 20-DOF pneumatic Bielefeld Shadow Hand. Our approach to grasping is based on a reach-pre-grasp-grasp scheme loosely motivated by human grasping. We comparatively describe the two robot setups, the control schemes, and the grasp type determination. We show that the grasp strategy can robustly cope with inaccurate control and object variation. We demonstrate that it can be ported among platforms with minor modifications. Grasping success is evaluated by comparative experiments performing a benchmark test on 21 everyday objects.

Real-time 3D segmentation of cluttered scenes for robot grasping

“…We used the segmentation approach for autonomous grasping with the 24-DOF Shadow Robot Hand employing our biologically inspired grasping strategy [7]. To obtain a coarse shape model of the object -which is required to select a grasp prototype, i.e.…”

“…While our previous work in this direction [7] was limited to simple 2D scenes, in this paper we propose a 3D scene segmentation approach which separates objects and provides coarse shape and pose information suitable for grasping. The algorithm is model free and employs only generic smoothness constraints.…”

Trying to Grasp a Sketch of a Brain for Grasping

“…We have found that already a very restricted set of five hand preshapes (all finger precision grasp, two finger precision grasp, power grasp, two-finger pinch grasp and three finger "special" grasp) is sufficient to achieve successful grasping for a wide range of common objects [RHSR07,Roet07] Grasping requires that each object is assigned to one hand preshape (this is currently not automated; the "correct" preshape for an object is commanded by a human operator making a corresponding hand gesture in front of a camera) and that the robot hand is positioned in a known distance and orientation to the target object. Therefore, object position and orientation has to be determined, e.g.…”

“…A tentative proposal within the EURON initiative is based on a bimanual Barrett hand system and proposes to evaluate grasp success for a number of (artificial) benchmark objects [Morales06]. A different benchmark, employing the set of 21 widely available household objects (shown in Fig.2), has been suggested in [Roet07] and has been used to compare grasp optimization schemes on two different robot hands [RHSR07].…”

Model-Based Evolution of a Fast Hybrid Fuzzy Adaptive Controller for a Pneumatic Muscle Actuator

“…On the one hand are their advantageous power/weight ratio, natural compliance, cleanness and structural simplicity, on the other are their nonlinear character, hysteresis, as well as other issues associated with pneumatic systems in general (non-negligible time delay and the need for bulky power sources). Nevertheless, a lot of effort has gone into PAM-based systems research corroborated by a number of various works, for example see [1][2][3][4][5]. There are basically two main areas for which various PAM applications have been proposed: robotics [6][7][8][9] and biomedical engineering [10][11][12].…”