On-Line Interactive Dexterous Grasping

Abstract

Abstract. In this paper we describe a system that combines human input and automatic grasp planning for controlling an artificial hand, with applications in the area of hand neuroprosthetics. We consider the case where a user attempts to grasp an object using a robotic hand, but has no direct control over the hand posture. An automated grasp planner searches for stable grasps of the target object and shapes the hand accordingly, allowing the user to successfully complete the task. We rely on two methods for achieving the computational rates required for effective user interaction: first, grasp planning is performed in a hand posture subspace of highly reduced dimensionality; second, our system uses real-time input provided by the human user, further simplifying the search for stable grasps to the point where solutions can be found at interactive rates. We demonstrate our approach on a number of different hand models and target objects, in both real and virtual environments.

The Columbia grasp database

“…The database is intended to be used in conjunction with GraspIt! or a similar simulation tool; as we have shown in previous work [10], [3], planning results obtained in simulation can be successfully applied to real robotic hands performing grasping tasks.…”

Interactive teaching of task-oriented robot grasps

“…The external wrench is assumed unknown, and therefore it is modeled as a unit ball to equally consider all forms of external wrench in all directions. As a result, neither [17] nor [18] address the issue of task-related grasping.…”

“…GraspIt! has integrated support for online real-time human interaction [17]. However, the grasp quality is computed with the standard GWS measures.…”

Synthesis of detailed hand manipulations using contact sampling

“…Common strategies such as controlled sliding and rolling contact [Tournassoud et al 1987;Cai and Roth 1987;Cole et al 1992;Cherif and Gupta 1999], or finger gait [Hong et al 1990;Han and Trinkle 1998;Xu et al 2007] can largely improve the capability of robotic manipulators. In addition to synthesis, robotics researchers can also transfer manipulation to different situations by analyzing the quality of contact locations [Pollard and Hodgins 2002;Pollard and Wolf 2004] and the relative motion between the object and the wrist [Ciocarlie and Allen 2008;Hamer et al 2011]. We draw many insights from the robotics literature in developing our method, but ours is fundamentally different in the underlying assumption.…”