Karun B. Shimoga

Robot grasp synthesis algorithms: a survey

This article presents a survey of the existing computational algorithms meant for achieving four important properties in autonomous multifingered robotic hands. The four properties are: dexterity, equilibrium, stability, and dynamic behavior The multifingered robotic hands must be controlled so as to possess these properties and hence be able to autonomously perform complex tasks in a way similar to human hands. Existing algorithms to achieve dexterity primarily involve solving an unconstrained linear programming problem where an objective function can be chosen to represent one or more of the currently known dexterity measures. Algorithms to achieve equilibrium also constitute solving a linear program ming problem wherein the positivity, friction, and joint torque constraints of all fingers are accounted for while optimizing the internal grasping forces. Stability algorithms aim at achiev ing positive definite grasp impedance matrices by solving for the required fingertip impedances. This problem reduces to a nonlinear programming problem. Dynamic behavior algorithms determine fingertip impedances, which, when achieved, lead to a desired dynamic behavior. This problem, too, becomes a linear programming problem. If a robotic hand has to acquire any or all of these proper ties, the corresponding algorithms should become integral parts of the hand control system. These algorithms are collectively referred to in this article as robot grasp synthesis algorithms.

A survey of perceptual feedback issues in dexterous telemanipulation. II. Finger touch feedback

For part I see ibid., p.263-70 (1994). The existing touch display technologies in the literature are surveyed. This survey indicates five main approaches to touch feedback, involving visual, pneumatic, vibro-tactile, electro-tactile and neuromuscular stimulations. A pneumatics approach could use air jets, air pockets or inflatable bladders to provide touch feedback cues to the operator. Similarly the vibro-tactile approach could use vibrating pins, voice coils, or piezoelectric crystals to provide tickling sensation to the human operators skin to signal the touch. The electro-tactile stimulation method can provide electric pulses of appropriate width and frequency to the skin, while the neuromuscular stimulation approach provides the signals directly to the primary cortex of the operators brain. With regard to this, 17 devices, most of which were built for sensory substitution purposes, are examined and compared for their suitability as touch feedback devices for dexterous telemanipulation.<<ETX>>

Soft robotic fingertips. Part I: a comparison of construction materials

Three potential problems exist in multifingered robotic hands. First, the impact forces that result during each instant of grasping a rigid object can affect the functioning of the fingertip sensors. Second, a hand with hard fingers cannot securely grasp objects that have uneven surfaces due to the poor conformability of the fingers. Third, repetitive strains are induced into the fingers throughout a manipulation task. If they are not dissipated, the manipulation becomes jerky, the func tioning of the fingertip sensors is affected, and the life of the fingers skeletal structure may become short. In this work, six fingertips constructed from plastic, rub ber, sponge, fine powder, paste, and gel are experimentally compared for their ability to overcome the above problems. Results show that the sponge fingertip is the most suitable and the plastic fingertip the least suitable for our application. For practical reasons, however, gel is preferred over sponge. In view of these results, it is recommended that future robotic hands employ soft fingers, or at least fingers with soft tips, constructed out of carefully chosen materials.

Soft materials for robotic fingers

Three potential problems exist in multifingered hands. The impact forces that result during each instant of grasping a rigid object can affect the functioning of the finger tip sensor. A hand with hard fingers cannot securely grasp objects that have uneven surfaces due to the poor conformability of the fingers. Repetitive strains are induced into the fingers throughout manipulation task. Carefully chosen materials-plastic, rubber sponge, a fine powder, a paste, and a gel-were experimentally compared for their ability to overcome these three problems. Results showed that sponge is the most suitable and plastic is the least suitable for the application. For practical reasons, however, the gel was a good compromise over the sponge. It is recommended that future robotic hands constitute a soft finger or at least fingers with soft tips, constructed out of carefully chosen materials.<<ETX>>

Touch and force reflection for telepresence surgery

This paper describes our on-going work on the development of an advanced telepresence glove, capable of providing touch and force reflection to the human operator. Such a glove will be essential in a telepresence surgery environment. The development of the glove currently constitutes three steps. First, an algorithm is being developed to map human finger motions on to the generally dissimilar robot hands, using the human grasp posture as the reference criteria. Second, a touch reflection system is being built where a set of touch sensors, placed on the robot finger tips, detect touch and activate a set of shape memory alloy based actuators attached to the human finger tips. Third, a force reflection system is being developed where a set of flexible pneumatic tubes, attached to the human fingers, reflect the grasping forces of the remote robot fingers.

Soft robotic fingertips. Part II: modeling and impedance regulation

Part I of this article showed that making robotic fingers from soft materials helps them cope with the following three prob lems. First, the impact forces that occur at the instant of grasping are reduced. Second, commonly used objects with uneven surfaces can be grasped comfortably due to the confor mal nature of the soft fingers. Third, the repeated strains that are induced during manipulations are dissipated. The problem to solve now is that of modeling and controlling the overall impedance of a soft finger or of a finger that has a soft tip. This is the issue addressed in this part. This part presents the experimental determination of the pas sive impedance parameters of a soft (gel-filled) fingertip and formulates an approach to account for the soft fingertip model when controlling the overall impedance of a finger that carries the soft tip. The approach is demonstrated via four experi ments, each with a unique goal. Also showed experimentally is how the presence of passive damping helps reduce the peak impact forces that occur as a rigid object is grasped by the fingers of a robotic hand.

Grasp admittance center: a concept and its implications

A grasp with an admittance center will have the following three consequential advantages: (1) the disturbing forces and motions experienced by the grasped object result in motions and forces on the respective degrees of freedom only; (2) the task of specifying the behavior of the object then reduces to that of specifying the behaviors of a finite number of linear second-order systems; and (3) each degree of freedom of the grasp will then be a linear second-order system and will always be stable. Since all such degrees of freedom are dynamically decoupled, the grasp as a whole will be stable. Contributing to achieve such a behavior in articulated multifingered grasps is the prime objective of the work.<<ETX>>

Constructing multifingered grasps to achieve admittance center

Extending the grasp compliance center concept to grasps in dynamic situations, previous work formulated a concept termed the grasp admittance center and showed that a grasp with an admittance center will have four distinct advantages: stability, decoupled force/motion relation, decoupled time response, and the ease of assigning the grasp dynamic behavior. The authors obtain the analytical conditions that must be satisfied for an admittance center to be achievable. They develop a procedure for computing the necessary fingertip impedances to achieve a desired admittance center. The results of experiments showing the regulation of the apparent impedances of a typical finger are presented.<<ETX>>

Visual and force feedback to aid neurosurgical probe insertion

The objective of this work is to develop a system for safe planning and execution of neurosurgery. The system under construction consists of a virtual model of the human brain, interfaced to a force-reflecting probe-guide insertion device. With such a system, neurosurgeons can plan safe paths for inserting biopsy needles and other surgical tools while avoiding possible rupture of blood vessels during the surgery. The operational principle as well as the design of the force and visual feedback system are presented in this paper.

Real-time computational aspects of multiple manipulator systems

In this paper, we address the topic of numerical computation in a system of multiple manipulators, one which has received scant attention despite a great deal of research in development of control schemes and the proliferation of similar work on single manipulator systems. Different approaches to computations in kinematics, statics and dynamics of multiple manipulator systems are studied and compared. The results show that significant savings in computational count can be achieved through customization of models. A systematic approach to customization is also described herein.