In this article, we characterize the passivity of a class of haptic systems modeled as a simple sampled-data system. We guarantee passivity by ensuring that there is sufficient damping in the haptic interface. Previous work established a necessary and sufficient bound on damping, but the corresponding mathematical expressions were complicated, and the derivation was not completely rigorous. After providing a rigorous proof, we derive a more tractable expression. Using this improved expression, we establish passivity conditions for several classes of transfer functions representing virtual environments, including some special cases with time delay. The original results assumed that the operator can be modeled by a passive but otherwise arbitrary transfer function. This assumption is weakened to allow the operator model to have a shortage of passivity. This requires only a slight modification of the passivity bound.

This paper presents the design of a bioinspired artificial fingertip that resembles the mechanical behavior of a human fingertip under conditions of both static deformation and high frequency excitation. The artificial fingertip is constructed around a deformable spherical membrane filled with a cellulose sponge, itself connected to a rigid structure that acts as a bone. Force-deformation characteristics and response to a transient mechanical perturbation are both shown to be in good qualitative agreement with those of a real finger. More importantly, the fingertip exhibits friction reduction when interacting with TPads (variable friction tactile displays based on transverse ultrasonic vibrations). Comparison with artificial fingertips that do not exhibit friction reduction suggests that mechanical damping characteristics play a key role in the amount of friction reduction achieved.

Texture modeling strives to encapsulate the important properties of texture in a concise representation for interpretation, storage, and rendering. Models for tactile texture have yet to describe a representation that is both perceptually complete and sufficiently compact. In this work, we take inspiration from models of visual and auditory texture and propose a spatial spectrogram representation of tactile texture that separates localized features from textural aspects using a windowed Fourier decomposition. We investigate the length scales at which humans can perceive localized features, and represent textures as spectrograms that capture those local features. Additionally, we demonstrate a reconstruction algorithm capable of recreating texture from a spectrogram representation with no perceptual consequence.

This work aims to demonstrate and explain a nearly century old electrostatic haptic effect in human fingertips, which has since gone unreported. This effect, based on the original work of Johnsen and Rahbek [1], as well as research on electrostatic chucking devices [2], is capable of producing electrostatic forces on the finger an order of magnitude greater than those previously reported in literature. It is also capable of working with DC excitation, an aspect which stands out against previous reports which utilize purely AC excitation. This work also proposes a unified force model for this effect, drawn from electrostatic chuck research, and resolves this model with those in previous reports. We briefly discuss the background and specifics of the Johnsen-Rahbek effect, and include measurements made with our own electroadhesive surface and experimental apparatus. Finally, we discuss how this model fits in with previous observations, and its implications going forward.

With the ubiquity of wearable computing, an important and emerging challenge is to understand how to design wearable information displays for non-visual, non-auditory interaction. This is particularly relevant to the design of accessible technologies for people with vision impairments. Working towards this aim, we developed a smartwatch prototype that uses variable friction surface haptics to test initial design concepts. Through interviews and iterative prototyping with seven blind users, we identified three key use cases for a haptic smartwatch as well as embodied conceptual models for presenting haptic information. We found that a physical clock face, compass, and numerical keypad are productive representations for presenting information haptically, and these models build on existing tactile and spatial understandings of our target user group.

The TPad Tablet combines an Android tablet with a variable friction haptic touch-screen and offers many novel interaction possibilities. For example, unique textures may be associated with different user interface elements, such as text boxes and buttons. This paper presents an Android AccessibilityService that was created to give operating system-wide (OS) access to haptic effects. Prior to this work, the haptic feedback of the TPad could be controlled only from within specific applications. With the new implementation, all applications and primary user interfaces (e.g. home screen) will have access to the TPad. Rather than focus on specific elements or applications, we seek to provide a high fidelity haptic experience that elevates the TPad's accessibility to the standard of Talkback and Voiceover, Android's and Apple's accessibility programs respectively. The code for the application is available on our website.

This paper explores the use of variable friction surface haptics enabled by the TPad Tablet to support affective communication between pairs of users. We introduce three haptic applications for the TPad Tablet (text messaging, image sharing, and virtual touch) and evaluate the applications with 24 users, including intimate couples and strangers. Participants used haptics to communicate literal texture, denote action within a scene, convey emotional information, highlight content, express and engage in physical playfulness, and to provide one's partner with an experience or sensation. We conclude that users readily associate haptics with emotional expression and that the intimacy of touch in the contexts we study is best suited for communications with close social partners.

This work investigates how haptic percepts are combined across two fingertips. Two single-degree-of-freedom haptic interfaces were used to present virtual bumps to the thumb and index finger of subjects' right hand. As subjects slid the two interfaces from left to right while maintaining a fixed finger separation, they would encounter one bump with the index finger and one with the thumb. The objective bump locations were varied randomly, from spatially coincident to separated by slightly more than the fingertips. Subjects were asked to report the number of bumps in the objective world and the number of times they encountered each bump, and also to point to the locations of the bumps. We found that subjects exhibited a strong bias toward reporting a single objective bump in the virtual world. However, the percept varied from one bump encountered twice, when the two virtual bumps were nearly spatially coincident, to one bump encountered once, which occurred when the two virtual bumps were close to finger separation and therefore encountered nearly simultaneously. The latter result is evidence for a kinesthetic grouping phenomenon: temporally synchronized sensations at multiple fingers can be perceptually collapsed into a single percept in both time and space.

The human fingertip is extremely sensitive to lateral (shear) forces that arise in surface exploration. We and others have developed haptic displays that work by modulating surface friction via electrostatic attraction. Despite the demonstrated ability of these displays to render haptic environments, an understanding of the fingertip-surface interface is lacking. We have developed a tribometer for measuring the lateral frictional forces on a fingertip under well-controlled conditions. We show an expected square law dependence of frictional force (and inferred electrostatic normal force) on actuation voltage, although we observe a large person to person variability. We model an expected dependence of the frictional force on the frequency of the actuation voltage, predicting a first order cut off below about 500Hz. However, our measurements are unambiguously at odds with the model's predictions.

Touch interactions have refreshed some of the 'glowing enthusiasm' of thirty years ago for direct manipulation interfaces. However, today's touch technologies, whose interactions are supported by graphics, sounds or crude clicks, have a tactile sameness and gaps in usability. We use a Large Area Tactile Pattern Display (LATPaD) to examine design possibilities and outcomes when touch interactions are enhanced with variable surface friction. In a series of four studies, we first confirm that variable friction gives significant performance advantages in low-level targeting activities. We then explore the design space of variable friction interface controls and assess user reactions. Most importantly, we demonstrate that variable friction can have a positive impact on the enjoyment, engagement and sense of realism experienced by users of touch interfaces.

This paper presents a framework that enables an operator to haptically and visually interact with a simulated dynamic environment subject to virtual holonomic constraints. The framework combines a geometric constraint solver with a constrained dynamics simulation engine that controls an admittance-type haptic display. This system takes on relevant issues in the context of assisted teleoperated tasks, from providing an intuitive interface for creating and combining virtual constraints, to haptically displaying rigid motion constraints in simulated environments subject to desired inertial dynamics. Two experiments carried out using the Cobotic Hand Controller haptic display are presented.

We propose a novel control method for lowerlimb assist that produces a virtual modification of the mechanical impedance of the human limbs. This effect is accomplished through the use of an exoskeleton that displays active impedance. The proposed method is aimed at improving the dynamic response of the human limbs, while preserving the user’s control authority. Our goal is to use active-impedance exoskeleton control to improve the user’s agility of motion, for example by reducing the average time needed to complete a movement. Our control method has been implemented in a 1-DOF exoskeleton designed to assist human subjects performing knee flexions and extensions. In this paper we discuss an initial study on the effect of negative exoskeleton damping (a particular case of active-impedance control) on the subject’s time to complete a target-reaching motion. Experimental results show this effect to be statistically significant. On average, subjects were able to reduce the time to complete the motion by 16%.

The haptic sense provides continuous information during physical human-machine interaction. Humans can respond very quickly and effectively to such feedback, such as a driver making rapid steering adjustments when his vehicle hits a pothole. It may be possible for designers to take advantage of this interaction by providing users with well- designed haptic communications to assist in manual control tasks. In this paper, we describe an experiment conducted to test this idea. Subjects were instructed to step on a modified stair climber at a consistent workout pace while simultaneously completing a mental distraction task. They were provided with one of five different haptic communications to assist in maintaining their workout intensity level. Results show that haptic communications helped subjects step as much as 1.7% faster with as much as 3.95% more power output than without the haptic communications, although none of the communications beneficially affected velocity or power standard deviations. Four of the five communications had no significant impact on mental task performance.

This paper investigates how collaborative robots (cobots) can assist a human by mechanically constraining motion to software-defined guide paths, and introduces simple and efficient tools to design ergonomic paths. Analysis of the movements of seven subjects with the Scooter cobot reveals significant differences between guided movements (GM) and free movements (FM). While FM requires learning for each novel task, movements in GM are satisfying from the first trial, require little effort, are faster, smoother, and with fewer back and forth corrections than in FM. Operators rely on path guidance to rotate the Scooter and direct it along curved trajectories. While these advantages demonstrate the strength of the cobot concept, they do not show how guide paths should be defined. We introduce tools to enable the cobot and its operator to collaboratively learn ergonomic guide paths and adapt to changes in the environment. By relying on the haptic sensing, vision, and planning capabilities of the human operator, we can avoid equipping the cobot with complex sensor processing. Experiments with human subjects demonstrate the efficiency and complementarity of these guide paths design tools

In the Cobotic Hand Controller, we have introduced an admittance display that can render very high impedances (up to its own structural stiffness). This is due to its use of infinitely variable transmissions. While admittance displays typically excel at rendering high impedances, the incorporation of infinitely variable transmissions in the Cobotic Hand Controller allows the stable display of a wide dynamic range, including low impedances. The existence of a display that excels at rendering high-impedance constraints, but has high-fidelity control of low impedances tangent to those constraints, has led us to describe an admittance control architecture not often examined in the haptics community. In this paper, we develop a comprehensive approach that enables rendering of rigid motion constraints while simultaneously preserving the physical integrity of the intended inertial dynamics tangent to those constraints. This is in contrast to conventional impedance-control algorithms that focus primarily on rendering reaction forces along contact normals with constraints. We present this algorithm here, which is general to all admittance displays, and report on its implementation with the Cobotic Hand Controller. We offer examples of rigid bodies and linkages subject to holonomic and/or nonholonomic constraints

We present experimental results of thrust produced by a robotic propulsor, the design of which is inspired by the ribbon fin of the South American black ghost knifefish (Apteronotus albifrons). This remarkably nimble fish moves by oscillating its ribbon fin rays out of phase and thereby passing a traveling wave along the fin's length. Combinations of thrust from the ribbon fin and body rolls produced by the two pectoral fins enable the black ghost to swim in nearly any direction without bending its body. The fish's agile locomotor system is tightly integrated with its omnidirectional, active sensing system. The robotic ribbon fin has eight individually actuated metal rays which are linked by a thin latex sheet. The experimental results demonstrate the effect of varying the propulsive wave's frequency, amplitude and length on the robotic fin's thrust production. We found that thrust production peaks at particular combinations of the three variables and that the fin could produce steady forward thrust, despite the relatively small number of rays. The robotic ribbon fin has potential application as a propulsor for future underwater vehicles, in addition to being a valuable scientific instrument in understanding the swimming mechanics of the black ghost and similar fish

In this paper we are concerned with allowing the operator of a haptic display to interact with virtual systems having significant inertial dynamics and realistic constraints. We review the mathematical structure arising from the kinetic energy metric, required to create a virtual dynamics simulation consisting of rigid-body dynamics along with holonomic and/or nonholonomic motion constraints. We develop an admittance controller composed of feedforward and feedback terms, while preserving the integrity of the intended virtual dynamics simulation. This controller is implemented on the Cobotic Hand Controller, an admittance-type haptic display, and two examples are discussed.

A novel method based on deformable length of elastic element control (DLEEC) to realize the stiffness display for perception of virtual soft object is proposed, and the stiffness display interface device has been developed and is presented. The stiffness display interface device is composed of a thin elastic beam and an actuator to adjust the length of the beam. The deformation of the beam under a force is proportional to the third power of the beam length. By controlling the beam length, the stiffness display device can reproduce the stiffness of the virtual object from very soft to hard, so that the human fingertip can feel it as if he directly touches with the virtual object by interacting with the device. A real time position control algorithm is employed to guarantee the real time stiffness display.

We consider teleoperation in which a slave manipulator, seen in one or more video images, is controlled by moving a master manipulandum. The operator must mentally transform (i.e., rotate, translate, scale, and/or deform) the desired motion of the slave image to determine the required motion at the master. Our goal is to make these mental transformations less taxing in order to decrease operator training time, improve task time/performance, and expand the pool of candidate operators. In this paper, we introduce a framework for describing the transformations required to use a particular teleoperation setup. We analyze in detail the mental transformations required in an interface consisting of one camera and display. We then expand our discussion to setups with multiple cameras/displays and discuss the results from an initial experiment.

A novel, six-degree-of-freedom active haptic device is introduced. A parallel kinematic design and the use of non-holonomic continuously variable transmissions provide extremely high stiffness in directions that would violate a virtual constraint. At the same time, smooth motion is permitted tangential to virtual constraints and in open space. High quality constraint surfaces having one to five dimensions can be displayed. A notable feature of this device is the mutual coupling of all six linear actuators to a common rotating cylinder, which can, optionally, be powered. The resulting mechanism is simple to control, and allows several new control strategies in Cobotic haptics.

This study investigates natural single-arm interaction with kinematic constraints. Smooth, frictionless, kinematic constraints reduce the degrees-of-freedom of motion at the hand, but add force freedoms. These force freedoms allow the hand to push and pull against the constraints with no effect on the task. Understanding how subjects take advantage of kinematic constraints will be useful in designing constraint surfaces for assisted manipulation. This paper reports the results of an experiment studying how subjects make use of the presence of a kinematic constraint in a static planar single-arm task. Subjects are asked to hold a handle that is free to slide on a linear rail, and to apply a force tangent to the rail to resist a pulling force. Thus the goal of the task is to hold the handle stationary. Subjects are also free to apply any force normal to the rail, as these forces have no effect on the task. This freedom does not exist without a kinematic constraint. We find that subjects make use of the force freedom by applying significant forces against the constraint in a consistent and constraint-configuration-dependent fashion. We show that the constraint forces can be predicted by a convex, scale-invariant objective function on the hand force space. The level curves of this objective function can be found directly from the experimental data without any biomechanical modeling.

Remote systems are essential for reducing risk to human workers from hazardous radiation and difficult work environments, while improving productivity and reducing costs. The major drawback of currently available remote manipulator systems is that teleoperation is slow and imprecise. The presented work focuses on enhancing remote operation of tools for D&D tasks by introducing teleautonomy and telecollaboration. In teleautonomy, the robot performs a given task autonomously, while the human operator intervenes in the process as a supervisor. In telecollaboration, the human operator is passively constrained by a virtual fixture, but is responsible for the motion. This work, sponsored by the US Department of Energy (DOE) Environmental Management Science Program (EMSP), builds on a reactive, agent-based control architecture and robot control technology.

Often two people must work together physically on a common task, such as lifting and positioning; a long board, or, in our model experimental system, turning a two-handled crank. Such tasks involve communication between the people, mediated by the task kinematics and dynamics: each person feels forces and motions produced by the other and derive some meaning from them. Tasks may include a degree of competition: the two people may not have exactly the same goal in mind, and must negotiate a compromise. Understanding human-human communication is important in designing robots for interaction with humans, and for robots that provide powered assistance for human-human tasks (such as physical therapy). In this paper we describe early experiments in human-human physical interaction, with a 1 dof robot included in order to give experimental access to the exchange of forces and motions between the people. We report on Fitts' law-like tasks, in which the two people cooperate to move a cursor to a common target, or to targets that do not completely overlap. Our results suggest that human-human physical communication may be a rich area of study.

We have been exploring the use of passive robotic mechanisms for the display of virtual surfaces. Cobots are one way of producing virtual surfaces using a passive mechanism. Unlike powered robots, the nonlinear dynamics of the passive mechanism (e.g., an arm) can be felt by the user as a spatially varying apparent inertia. This effect occurs in many passive designs, including but not limited to cobots. We explain the variable apparent inertia as the projection of the spatially-varying inertia matrix onto the direction of motion, and discuss several ways to control the apparent inertia. We explore apparent inertia in detail for the unicycle two link arm, a cobot we have developed for experiments in single-arm motor control studies and rehabilitation. Special paths ("iso-mass contours") are found for this mechanism along which the apparent inertia is constant.

Passivity theory has been used for the past decade to derive stability conditions for human/machine interface applications. Demonstrating passivity of the haptic display implies stable and safe interaction for the human user. At the heart of the stability analysis is the physical dissipation provided by the haptic device, as it plays a key role in the design process for all components. This paper will derive the condition that the haptic device must satisfy in order to achieve passivity of the haptic display. These results will be used to investigate a general nonlinear device model.

The state-of-the-art in intelligent assist devices (IADs) is reviewed. IADs are computer-controlled tools that enable production workers to lift, move and position payloads quickly, accurately, and safely. Several examples of industrial applications are given, illustrating typical configurations and functionality, including strength amplification and virtual surfaces. The concept of human intent sensing is introduced and discussed, as are IAD safety and control considerations.

Cobots are devices for human/robot interaction, in which axes of motion are coupled to one another by computer-controlled continuously variable transmissions rather than individually driven by servomotors. We have recently built a cobot with a three-dimensional workspace and a 3-revolute parallelogram-type mechanism. Here we present the control methods for the display of virtual surfaces and for free mode in which the cobot endpoint moves as if it were unconstrained. We provide experimental results on the performance of the free, virtual path, and virtual surface controllers.

Cobots are devices for human/robot interaction, in which axes of motion are coupled to one another by computer-controlled continuously variable transmissions rather than individually driven by servomotors. We have recently built a 3R parallelogram cobot. Here we present a control method for the display of virtual surfaces and for free-mode in which the cobot endpoint moves as if it were unconstrained. We show experimental results on the performance of the cobot in these modes.

Of several possible forms of human-robot collaborative manipulation, we focus on the case where the human and the robot jointly manipulate a common load. In our formulation, the robot's role is to provide a constraint surface to guide the motion of the load. The value of this form of interaction, in terms of ergonomics, accuracy, or speed, depends on how humans make use of such constraints. We are studying natural single-arm manipulation of a load constrained to move along a guide rail. In this paper we present results of experiments showing that subjects apply significant forces against the rail, depending on the configuration of the arm and the orientation of the rail. These forces are unnecessary for the manipulation task, and we hypothesize that humans apply forces against the constraint to simplify the manipulation task.

A general framework is presented for the design and analysis of cobot controllers. Cobots are inherently passive robots intended for direct collaborative work with a human operator. While a human applies forces and moments, the controller guides motion by tuning the cobot's set of continuously variable transmissions. In this paper, a path-following controller is developed that steers the cobot so as to asymptotically approach and follow a preplanned path. The controller is based on feedback linearization. Generality across cobot architectures is assured by designing the controller in task space and developing transformations between each of four spaces: task space, joint space, a set of coupling spaces, and steering space.

We describe a new robot architecture: the collaborative robot, or cobot. Cobots are intended for direct physical interaction with a human operator. The cobot can create smooth, strong virtual surfaces and other haptic effects within a shared human/cobot workspace. The kinematic properties of cobots differ markedly from those of robots. Most significantly, cobots have only one mechanical degree of freedom, regardless of their taskspace dimensionality. The instantaneous direction of motion associated with this single degree of freedom is actively servo-controlled, or steered, within the higher dimensional taskspace. This paper explains the kinematics of cobots and the continuously variable transmissions (CVT) that are essential to them. Powered cobots are introduced, made possible by a parallel interconnection of the CVT. We discuss the relation of cobots to conventionally actuated robots and to nonholonomic robots. Several cobots in design, prototype, or industrial testbed settings illustrate the concepts discussed.

Investigates the influence environment delay has on haptic systems. Work presented by the authors (1999) demonstrated that it is possible to account for environment delay (even though it is a non-passive behavior) and still achieve a stability result based on passivity concepts. The details were given for haptic systems with environments having impedance causality. This paper presents the companion results for haptic systems with environments having admittance causality, resulting in a slightly more complicated analysis. Although the main focus is on delay, the approach is valid for analysing any environment that exhibits non-passive behavior.

Design of haptic systems that guarantee stable interaction is a challenging task. Virtual environments are typically highly nonlinear-resulting in a nonpassive discrete-time model. This paper will investigate how nonlinear mass/spring/damper virtual environments can be designed to guarantee the absence of oscillations and other chaotic behavior in the signal presented to the human operator. In particular, delayed and nondelayed implementation of the mass/spring/damper virtual environment is considered, revealing a nonintuitive result with regard to the allowable local stiffness.

Intelligent assist devices (IADs) are a new class of hybrid devices for direct, physical, interaction with a human operator in shared workspaces. These devices-designed for the assembly line worker-can reduce ergonomics concerns that arise due to on-the-job physical and cognitive loading, while improving safety, quality and productivity. Cobots, a sub-set of IADs, implement software defined virtual guiding surfaces while providing some amplification of human power. They exemplify the central theme of this paper-that humans are critical in many assembly operations and ergonomics tools that enable them to perform their duties are necessary. The paper describes broad design principles for human-machine interaction in these industrial settings. Prototype industrial cobots designed for testing and validation are described.

A general framework is presented for the design and analysis of cobot controllers. Cobots are inherently passive robots intended for direct collaborative work with a human operator. While a human applies forces and moments, the controller guides motion by tuning the cobot's set of continuously variable transmissions. In this paper, a path following controller is developed that steers the cobot so as to asymptotically approach and follow a pre-planned path. The controller is based on feedback linearization. Generality across cobot architectures is assured by designing the controller in task space and developing transformations between each of four spaces: task space, joint space, a set of coupling spaces, and steering space.

Conventional force sensors are overdesigned for use in measuring human force inputs, such as is needed in research and application of human/robot interaction. A new type of force sensor is introduced that is suited to human-robot interaction. This sensor is based on optoelectronic measurements rather than strain gauges. Criteria for material selection and dimensioning are given, and results for linearity, noise, and drift are reported.

This paper derives conditions for the absence of oscillations for a parametrized class of nonlinear environments. This class includes discrete-time environments that can exhibit non-passive behavior. The motivation for considering non-passive discrete-time environments is based on the fact that an interesting class of passive continuous-time environments have non-passive discrete-time counterparts. A design methodology is introduced that provides relationships between the haptic device, virtual coupling and maximum negative stiffness exhibited by the environment.

Cobots are capable of producing virtual surfaces of high quality, using mechanical transmission elements as their basic element in place of conventional motors. Most cobots built to date have used steerable wheels as their transmission elements. We describe how continuously variable transmissions (CVTs) can be used in this capacity for a cobot with revolute joints. The design of an "arm-like" cobot with a three-dimensional workspace is described . This cobot can implement virtual surfaces and other effects in a spherical workspace approximately 1.5 meters in diameter. Novel elements of this cobot include the use of a power disk that couples three CVTs directly.

Cobots are a class of mechanically passive robotic devices, intended for direct physical collaboration with a human operator. The operator supplies all motive power while the cobot enforces software-defined guiding surfaces, or constraints. Cobots are intrinsically passive, safe devices. This is because, rather than employ powered actuators to produce constraint forces, cobots use "steerable" nonholonomic joints. Constraint forces are mechanical in origin, yet software defined. The simplest possible cobot is a unicycle which is steered by a servo system acting under computer control, but which is moved by a human operator. The unicycle cobot requires essentially no consideration of kinematics. Two fundamental control modes of the unicycle cobot, "virtual caster" and "constraint tracking", are reviewed. More complicated cobots, such as the three-wheeled "Scooter", require a set of kinematic tranformations relating configuration space to joint space. These transformations play a role in cobot control like that of the Jacobian in robot control.

Conventional approaches to haptic interface rely on high gain servos to implement virtual constraints. The role of the servo is to reduce the apparent degrees of freedom in such a way as to effectively constrain a human operator's motion. A significant drawback of this approach, however, is that the operator must interact directly with a high power system that is not inherently passive, and which may become unstable. In this paper, we present a novel approach to haptic display which allows virtual constraints to be implemented in a manner that is completely passive and therefore intrinsically safe. The key idea is to begin with a device having zero or one degree of freedom, and to use feedback control to increase the apparent degrees of freedom as necessary. This becomes possible with the use of nonholonomic joints, which have fewer degrees of freedom than generalized coordinates. The design and feedback control of several "programmable constraint machines" (PCMs) of this type are discussed.

Describes a new architecture for passive robots and haptic displays, which the authors call a programmable constraint machine (PCM). An n-DOF PCM can, under computer control, exhibit constraints (smooth, impenetrable virtual surfaces of dimensionality

Our group is interested in using haptic display for training tool use. Applications include training doctors to use tools during surgery, and training astronauts to use tools during EVA. This paper describes some of the challenges of creating realistic haptic perceptions of tool use. Many of these challenges stem from the importance of unilateral constraints during tool use. Unilateral constraints occur whenever rigid bodies collide, resisting the interpenetration of the bodies, but not holding the bodies together. To identify unilateral constraints, a tool/environment simulation must perform collision detection. To respond properly to a collision, the simulation must estimate the forces that ensue, and integrate the equations of motion. All of these computations must occur in real time, and the simulation as a whole must be stable (to ensure the user's safety). Approaches to these problems are described.

This paper addresses the performance of force-reflecting interfaces ("haptic displays"). The authors suggest that an important measure of performance is the dynamic range of achievable impedances-"Z-Width"-and that an impedance is achievable if it satisfies a robustness property such as passivity. Several factors affecting Z-Width-sample-and-hold, inherent interface dynamics, displacement sensor quantization, and velocity filtering-are discussed. A set of experiments designed to evaluate these factors is described and experimental results are presented. A striking result is that inherent interface damping exerts an overwhelming influence on Z-Width.>

When a human operator performs a task via a bilateral manipulator, the "feel" of the task is embodied in the mechanical impedance of the manipulator. Traditionally, a bilateral manipulator is designed for transparency; i.e. so that the impedance reflected through the manipulator closely approximates that of the task. "Impedance shaping bilateral control", introduced here, differs in that it treats the bilateral manipulator as a means of constructively altering the impedance of a task. This concept is particularly valuable if the characteristic dimensions (e.g. force, length, time) of the task impedance are very different from those of the human limb. It is shown that a general form of impedance shaping control consists of a conventional power-scaling bilateral controller augmented with a real-time interactive task simulation (i.e. a virtual environment). An approach to impedance shaping based on kinematic similarity between tasks of different scale is introduced and illustrated with an example. It is shown that an important consideration in impedance shaping controller design is robustness; i.e. guaranteeing the stability of the operator/manipulator/task system. A general condition for the robustness of a bilateral manipulator is derived. This condition is based on the structured singular value ( mu ). An example of robust impedance shaping bilateral control is presented and discussed.>

A power and impedance scaling bilateral manipulator (such as an extender or a macro-micro bilateral manipulator) can greatly enhance the manual capabilities of a human operator, but it can also compromise the inherent stability of the operator. A condition for the robust stability of an operator/bilateral manipulator environment system is derived using the structured singular value. The application of this condition is illustrated with several examples of power and impedance scaling via a two-channel bilateral manipulator.>

A form of impedance control is proposed which addresses the issue of stability and provides for the programming of robots for interactive tasks. Passive mechanical elements, specifically springs and dampers, are incorporated into the drive of a robot. The impedance of the robot is programmable because the stiffness and damping coefficients of these mechanical elements are themselves programmable. The design of these elements and their use are outlined. Prototype springs and dampers have been fabricated, and their performance in theory and in practice is discussed.>

It is noted that the performance of force-reflecting hand controllers, also called manual interfaces or manipulanda, is at present limited more by mechanical design than by computing hardware. Some of the important mechanical features of a high-performance manual interface are low inertia, low friction, high stiffness, backdriveability, high output force capability, multiple degrees-of-freedom, and sizable range of motion. A four-degree-of-freedom manipulandum that utilizes a three-degree-of-freedom parallel-link mechanism was designed to incorporate these features. The kinematic design of the manipulator prompted the definition of a force/torque workspace as the volume of operation within which certain maximum desired endpoint forces and torques can be achieved, given actuators of limited output. The design requires low actuator torques, is small in size, and does not approach joint limits in the workspace.>

A passive wrist, of fixed design, can be programmed to execute a wide range of useful control laws. Considered in particular are wrists whose actuators are unpowered hydraulic cylinders, the ports of which are coupled to one another via variable-conductance constrictions. The wrist is programmed by selection of these conductances, much as an analog computer is programmed. The range of control laws such a device can compute is characterized mathematically.>

The authors explore the sources of and solutions to robot contact instability associated with force feedback. The behavior of several linear robot models is analyzed in terms of the properties of their admittances. Using the novel technique of passive physical equivalents, an explanation for the often-observed instability of a force-controlled robot contacting a stiff surface is offered, and it is shown that a fundamental limit exists to the efficacy of any force feedback controller implemented on a robot with noncolocated actuators and sensors. Suggestions for improved force control involving both mechanical design and compensator design are also presented.>