Grasp database based on the presssure maps of robotic gripper: a preview

Building a grasp database to identify stable hand configuration for grasping a novel object is extremely useful in robotics community, and several databases are available in the literature for this purpose. In this paper, we briefly review the grasp databases that are available in the literature and provide an overview of the grasp database that we intend to build for stable grasp identification. The proposed database will differ markedly from the present ones because we account for the contact pressure maps while grasping and evaluate the grasp in real-time based on the contact force and the contact area. In this paper, we also report the contact pressure maps of daily household objects such as bottles and cans while grasping, and show that each recorded pressure map captures the underlying deformation at the contact, the material properties of contacting surfaces and display the influence of the physical characteristics of object on the contact formation. Specifically, contact pressure maps of empty bottle as well as fluid-filled bottle are explored to see their underlying contact deformation while grasping. A FE-based grasp analysis tool is suggested to evaluate the grasp stability in a constrained simulation environment using these maps.

[1]  David V. Hutton,et al.  Fundamentals of Finite Element Analysis , 2003 .

[2]  Matei T. Ciocarlie,et al.  The Columbia grasp database , 2009, 2009 IEEE International Conference on Robotics and Automation.

[3]  Xinyu Liu,et al.  Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics , 2017, Robotics: Science and Systems.

[4]  John F. Canny,et al.  Planning optimal grasps , 1992, Proceedings 1992 IEEE International Conference on Robotics and Automation.

[5]  Siddhartha S. Srinivasa,et al.  Benchmarking in Manipulation Research: Using the Yale-CMU-Berkeley Object and Model Set , 2015, IEEE Robotics & Automation Magazine.

[6]  Stefan Ulbrich,et al.  OpenGRASP: A Toolkit for Robot Grasping Simulation , 2010, SIMPAR.

[7]  Jimmy A. Jørgensen,et al.  VisGraB: A benchmark for vision-based grasping , 2012, Paladyn J. Behav. Robotics.

[8]  P. Abbeel,et al.  Yale-CMU-Berkeley dataset for robotic manipulation research , 2017, Int. J. Robotics Res..

[9]  Mathieu Aubry,et al.  Dex-Net 1.0: A cloud-based network of 3D objects for robust grasp planning using a Multi-Armed Bandit model with correlated rewards , 2016, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[10]  Matei T. Ciocarlie,et al.  Data-driven grasping with partial sensor data , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.