Complex manipulation with a simple robotic hand through contact breaking and caging

Abstract

Humans use all surfaces of the hand for contact-rich manipulation. Robot hands, in contrast, typically use only the fingertips, which can limit dexterity. In this work, we leveraged a potential energy–based whole-hand manipulation model, which does not depend on contact wrench modeling like traditional approaches, to design a robotic manipulator. Inspired by robotic caging grasps and the high levels of dexterity observed in human manipulation, a metric was developed and used in conjunction with the manipulation model to design a two-fingered dexterous hand, the Model W. This was accomplished by simulating all planar finger topologies composed of open kinematic chains of up to three serial revolute and prismatic joints, forming symmetric two-fingered hands, and evaluating their performance according to the metric. We present the best design, an unconventional robot hand capable of performing continuous object reorientation, as well as repeatedly alternating between power and pinch grasps—two contact-rich skills that have often eluded robotic hands—and we experimentally characterize the hand’s manipulation capability. This hand realizes manipulation motions reminiscent of thumb–index finger manipulative movement in humans, and its topology provides the foundation for a general-purpose dexterous robot hand.

Designing Underactuated Graspers with Dynamically Variable Geometry Using Potential Energy Map Based Analysis

“…By solving (1) at each object-grasper configuration, we can predict the gross motion of the object by following potential energy gradients from high potential energy to low potential energy configurations. This was experimentally shown in [7], [10].…”

“…For a given object O ∈ O, position p ∈ P j , grasper design g i j , and actuation command u i j ∈ U i , we can calculate the set of scaled contact wrenches, c i pj × Λ i pj ∈ R Nc×3 and its corresponding convex hull H i pj . Note that we scale the contact wrenches by the caging score to ensure safe manipulation [10]. Repeating this procedure for each grasper (g i 1 → g i di ) and every actuation command in U i we can construct the overall convex hull, H i p , given the set of all contact wrenches at p for a specific object,…”

“…By understanding the variation of potential energy across different configurations, an object can be driven to desired states. The validity of the method has been shown in simulation via resultant force fields [22] and in physical experiments where friction was kept intentionally low [10]. The computational cost of generating and utilizing potential energy maps for design and planning is low and their functional impact is high.…”

“…Caging grasps are important for secure transport during pick-and-place tasks, since the object can move with the robot even in the absence of large contact forces [9]. Additionally, Bircher et al used a fully-actuated grasper to show that caging grasps are essential for performing safe inhand manipulation [10]. Given their significance, designing graspers geared toward caging is a promising direction.…”

“…Prehensile manipulation could be performed using the derived relationship between actuator commands and the resulting object-grasper configuration [17], [18]. The concept of potential energy maps, which describes how the energy of the object-grasper system fluctuates throughout the system configuration space, is an encouraging method for intelligent manipulation, applicable to both underactuated and fully-actuated graspers [19], [20], [8], [21], [10]. By understanding the variation of potential energy across different configurations, an object can be driven to desired states.…”