Haptic Screen
Iwata Lab

This is a very interesting device. Basically this is very similar to the "pin cushion" discussed previously by the group. They use 5 ball screws attached to a rubber skin. The result is a 20cm x 20cm deformable surface. Hard to see how you would represent anything other than bumpy surfaces with that implementation, but if you added more ball screws you could make a very bumpy surface and perhaps even represtent a model being pushed through the rubber.

reference missing.
I emailed several people in the Iwata lab and requested a reference for an English version of their paper. (4/4/0)

Tactile Shape Displays for Small Scale Shape Feedback
Harvard
William Peine, Parris Wellman, and Robert Howe

This is a linear pin display made up of a single row of 10 individually actuated pins. They state that they used the single line since the display was to be moved across the surface of the finger pad anyway. However, they also acknowlege that the choice also simplifies the design. Each pin is controled by a SMA wire that shortens under current, thus pushing the pin upward.

Wellman, P.S., Peine, W.J., and Howe, R.D., 1997, "Mechanical Design and Control of a High-Bandwidth Shape Memory Alloy Tactile Display," in Proc. International Symposium of Experimental Robotics, (Barcelona, Spain).

Elastic Force Sensor
Iwata Lab

Another nice idea from the Iwata Lab. This one is like a cross section of a rubber hose. The hose can be deformed to various shapes, all similar to a cylinder of course. The idea seems to be that the user holds the device and the device takes on the shape of the model in a local area. No real info at their site and like their other pages the paper is in Japanese.

reference missing.
I emailed several people in the Iwata lab and requested a reference for an English version of their paper. (4/4/0)

Sarcos Dextrous Arm
Sarcos Inc.
Steve Jacobsen, F. Smith, E. Iverson, and D. Backman

The Sarcos Dextrous master is a high inertia device with 10 degrees-of-freedom and a complex dynamics structure. While it is worn around the users arm, and is antropmorphic, it is not an exoskelaton. It is a grounded device that is capabale of producing external forces.

Jacobsen, S.C., Smith, F.M., Iversen, E.K., and Backman, D.K., 1990, "High Performance, High Dexterity, Force Reflective Teleoperator," in Proc. 38th Conf. Remote Systems Technology, (Washington, D.C.), pp. 180-185.

Phantom
Sensable Technologies, Inc.
Thomas Massie and Kent Salisbury

This is the standard device used by most. It is 6 dof input but only 3 dof output. They are working on a device that is 6 dof output, but it is on their back burner apparently. Various sizes of the device are available. Their small desktop has a 16cm x 13cm x 13cm workspace. A max force of 6.4N can be driven with nominal position resolution of 0.02mm. The device has a stiffness of 3.16N/mm with very low inertia of under 75g. Their largest device, the 3.0, has a workspace of 42cm x 59cm x 82cm, max force of 22N, nominal resolution of 0.02mm, stiffness of 1N/mm, and an inertia of under 150g. The version we at Utah use is the 1.5 model. It has a 19.5cm x 27cm x 37.5cm workspace, max force of 8.5N, nominal resolution of 0.03mm, stiffness of 3.5N/mm, and an inertia of under 75g. THe device produces very smooth and crisp results.

Massie, T.M. and Salisbury, J.K., 1994, "The PHANToM Haptic Interface: A Device for Probing Virtual Objects," in 3rd Annual Symp. Haptic Interfaces for Virtual Environment and Teleoperator Systems, (Chicago, IL), DSC-Vol 1, pp. 295-301.

Palmtop Display for Dextrous Manipulation
ATR Communication Systems
Haruo NOMA , Tsutomu MIYASATO and Fumio KISHINO

A haptic device for use in their Virtual Space Teleconferencing system. The device itself is a linkage much like the phantom. The end-effector however is not a pen, but is a palmtop display device. Manipulation of the device takes both hands, one on either side, with the visual display showing a local view of the scene to allow the user to look where he is working instead of looking at the large projection. Ideally, this allows for less of a need to hand-eye coordination and the awkwardness of manipulating an object in space from a distance. Another reason for the local display is that since the haptic device lies between the user and the projection display there is a depth conflict. This can ruin the stereo sensation and cause the user to drop out of emmersion. They state in their paper "Almost everyone who has tried our trial VSTC has complained about the lack of haptic sensation, so it is assumed that these factors are important for generating a realistic manipulating environment."

Noma, Kitamura, Miyasato, Kishino, 1996, "Haptic and Visual Feedback for Manipulation Aid in a virtual Environment", ASME-DSC-Vol.58, pp.469-476.

WYSIWYF Display
Carnegie Mellon University
Yasuyoshi Yokokohji, Ralph Hollis, and Takeo Kanade

This work uses a Puma robot as its haptic device. A portable TFT display is placed in between the user and the device. On the backplane of the displace there is a color CCD camera used to track the devices location which has several markers placed on it to aid this process. The objects have grab-points attached to them in order to match the physical grab-point that is the devices end-effector. This allows the models to be manipulated through these artificial grasp areas. One point of interest is that the device, and the background, are all painted blue so that the users hand can be extracted using Chroma Keying techniques and placed in the visual display instead of a graphical version being generated.

Y. Yokokohji, R. L. Hollis, and T. Kanade, 1996, "What you can see is what you can feel. -Development of a visual/haptic interface to virtual environment-," IEEE Virtual Reality Annual International Symposium, pp. 46-53.

Freedom 6S
MPB Technologies

This device is a competitor to the Phantom with a workspace of 22cm x 24cm x 22cm. unlike the Phantom this is a 6dof device, both input and output. It doesnt map to one model of the Phantom though. It has a max force of 2.5N and 125mNm torque. Its resolution is the same as the Phantom at 0.02mm. The stiffness is around 1.3N/mm and the inertia of th etip is 150g. Like the Phantom it has a 1kHz update rate. They also make a 3dof device, which is not the same device without torque sensors. Their web page did nt give specifications but did supply an image. This device is not at all like the Phantom ot the Freedom 6S. It appears to use three orthogonal tracks that the end-effector, located in the center, must folow along on.

reference missing.
how do you reference such devices from industry?
MPB Technologies Inc., Pointe Claire, QC, Canada.

UTAH/MIT Dextrous Hand
Sarcos Inc.

The web site offers very little information other than it is the "most advanced robotic hand developed to date." It was designed and manufactured through a colaboration between Sarcos Inc., the Center of Engineering Design at Utah and the Artificial Intelligence Lab at MIT.

reference missing.
john hollerbach probably can get a reference for this.

Rutgers Master II
Rutgers University
Daniel Gomez

Basically, this is a device that you wear on one hand. Pneumatic pistons are positioned in the palm such that when they are activated they apply a force to the first three fingers of the hand as well as the thumb. This allows the simulation of grasping forces.

reference missing.
i believe it is published work in vrais 96 or 95 but dont have the proceedings handy.

CyberGrasp
Virtex Inc.

A cable driven haptic glove from the same people that brought you the CyberGlove and CyberTouch (which we have here at Utah). They make both right and left gloves. To quote them "a revolution in human-computer-interface technology by enabling CyberGlove users to actually 'touch' computer-generated objects and experience realistic force feedback via the most natural interface possible - the human hand." Force is aplied to each finger, but only in the grasp direction, meaning no force can be applied to make the finger move toward the palm. This is simply the case since the force applied is accomplished by pulling a cable over a track system that ends at the finger tip. The say the glove is "unencumbering", but from the look of it I cant imagine it being easy to get into, or calibrate. This glove was developed under STTR contract to the US Navy for telerobotic applications. It has a maximum force of 12N per finger, there is a 12bit force resolution, it weighs 350g, the workspace is one meter radius hemisphere. In order to operate you must allready own a 22-sensor CyberGlove (18-sensor upgrades available) and you need a 6dof tracker for hand location and orientation.

reference missing.
how do you reference such devices from industry?
Virtual Technologies, Inc., Palo Alto, California.

Haptic Interfaces
University of Colorado
Dale Lawrence, Lucy Pao, Anne Dougherty, Stephen Wallace and Shane Brown

This is an interesting device because it is configurable from 3dof to 6dof. When at 3dof it resembles a pyramid with three rods connecting at a point where the user holds. When configured for 6dof there are three rods attached to each end of a stylus, similar to a Stewart Platform. The give the accuracy of the device to be similar to that of the Phantom with comparable max force and force resolution. They state "Motor and rod are smoothly coupled by preloaded rolling elements, eliminating backlash and cogging, yet providing an extremely stiff link which enables high bandwidth transmission of forces to the user's fingers."

D. A. Lawrence, L. Y. Pao, M. A. Salada, A. M. Dougherty, 1996, "Quantitative Experimental Analysis of Transparency and Stability in Haptic Interfaces," Proc. ASME Dynamic Systems and Control Division, (Atlanta, GA), DSC-Vol. 58, pp. 441-449.

Haptic Master
Iwata Lab

A 6dof desktop device based on the Stewart Platform. The stress of the design seems to be that it is very inexpensive, basically three pantagraphs in parallel. He states advantages of the design as its ability to carry large payloads and its compactness, but says that the small working volume and its inability to backdrive (reduction of friction) are certainly disadvantagous.

reference missing.
this device is commercial, but the company page is missing.

Magnetic Levitation Haptic Interfaces
Carnegie Melon
Peter Berkelman and Ralph Hollis

A haptic interface device based on the Lorentz magnetic levitation. The user grasps a levitated tool handle in order to interact with a virtual environment. This is a 6dof device, with one moving part. There is no friction in the device since there is no linkage, meaning noncontact actuation and sensing. The workspace seems small, at only 25mm translation, but I think it is to be used more like a joystick than a tracker. It has a high maximum force of 55N and torque of 6Nm. The position is rather accurate, like the Phantom at 0.05mm with a stiffness of 25N/mm. While they call the device compact because it only requires a deskside case, it seems that this also limits the device in that it must be placed off to the side of the user unless a giant hole is to be cut into the desktop.

P. J. Berkelman, Z. J. Butler, and R. L. Hollis, 1996, "Design of a Hemispherical Magnetic Levitation Haptic Interface Device," 1996 ASME IMECE, (Atlanta, GA), DSC-Vol. 58, pp. 483-488.

Pantograph
McGill University
Dinesh Pai and Vincent Hayward

Very simple planar device that can be used to apply forces to a device like a mouse. They offer three versions that seem to go from small to smallest. The smallest version is hand held and therefore a tracker needs to be used to locate where in the virtual space the operator is. In either case the device only applies planar forces and can therefore display a sculptured curve but not a surface. That is to say, not unless the surface is displayed one planar intersection curve at a time.

Ramstein, C. Hayward, V., 1994, "The Pantograph: a large workspace haptic device for a multi-modal Human-computer interaction. CHI'94, Conference on Human Factors in Computing Systems ACM/SIGCHI, (Boston, MA).

Excalibur - The Linear Haptic Display
University of Washington
Manuel Moreyra

This is another planar device that operates on a much larger area. The device is laid out on a large square surface with guide rails along each edge. The position of the device is at the intersection of two cross bars. At this intersection a third axis allows vertical movement. This gives rather high accuracy for the tracking of the end-effector. The workspace is 30cm x 30cm x 20cm, the max force is 100N and the nominal resolution is 0.008mm.

R. Adams, B. Hannaford, 1999, "Excalibur, A Three-Axis Force Display," ASME Winter Annual Meeting Haptics Symposium, (Nashville, TN)..

Immersion TouchSense
Immersion Corporation

They offer various haptic devices, but seem targeted mostly at the gaming market. They apparently have a six dof hand device, but I lacked the proper plug-ins to really investigate their web pages. They also advertise a "Laparoscopic Impulse Engine." From their web page you can see that they have several patents, develope chips for "simulating feel sensations", and also deal in the software side by producing foundation classes.

reference missing.

Pneumatic Haptic Interface
Southern Methodist University
Yildirim Hurmuzlu

The device they have built is a unilateral aluminum manipulator with four revolute joints. It tracks at the shoulder and elbow with one end fixed to the chair the user sits in and the other having a handle the user graps. The device is strapped to the operator by a collection of inflatable cushions. Force feedback is relized by actuating the joints by the pnematic cylinders. The system is contained in two processes run on a basic sgi and macintosh. The model complexity looks to be low, but isnt discussed in any depth.

Hurmuzlu Y., Ephanov A., and Stoianovici D., "Effect of a Pneumatically Driven Haptic Interface on the Perceptional Capabilities of Human Operators", to appear in Presence, MIT. Press.

Cobot
Northwestern University
Edward Colgate and Michael Peshkin

A cobot is a robot for direct physical interaction with a human operator, within a shared workspace. These devices seem to be radically different than the devices built by others. One of the devices produces its force feedback via a unicycle wheel. Another uses three wheels in order to produce twist force anolge with planar forces. They also show some work on "tetrahedral cvt" or continuously variable transmission. Here angular velocities of two shafts are remapped to an angle for the a steering roller. It is sadi to be the analog of a rolling wheel for their scooter type cobots. This type could be used as a revolute joint in more arm like devices.

R. Brent Gillespie, J. Edward Colgate, Michael Peshkin, 1999, "A General Framework for Cobot Control," 1999 International Conference on Robotics and Automation. (Detroit, MI).

Fingertip Haptic Display
University of Washington
Steve Venema and Elger Matthes

The goal here is to develope a small device that will apply haptic feedback to a single finger. They have built this device but have little results as of yet on their page. They do mention that their system uses a 16-bit DOS development environment on their Pentium class machines. They get the advantage of guaranteed updates and then all the obvious disadvantages. They are looking into using Solaris and then using Posix threads to bind jobs to different processors.

reference missing.

The Pen Based Force Display
University of Washington
Pietro Buttolo and Blake Hannaford

This is a very small device designed to be used with either a single finger tip of the tip of a pen. It is 3 dof devicw with two of these being used for position (a rocker like device) and the third being able to simulate a button. They describe one of its uses to be a substitute for a mouse.

P. Buttolo, B. Hannaford, 1995, "Pen Based Force Display for Precision Manipulation of Virtual Environments," in Proc. VRAIS-95, (Raleigh, NC), pp. 217-225.

3D Pen
Iwata Lab

No real information on the device other than an image and a bad mpeg movie. The device attaches the tip of a pen orthoganally to a cross beam. two devices like Phantoms are then connected to the ends of this cross beam. The result, I think, is a 5dof device since roll cannot be reproduced.

reference missing.

Hand Held Force Display
Iwata Lab

This device is a simple four-bar linkage, much like the Phantom, but is small and attaches to the forearm of the user. Again, no text on the device so this description comes from the animated GIF they have online. It seems to me that the device is a 2dof planar device with a very small workspace. Interesting mount to the arm though, similar to what I had suggested for the pin cushion mounting.

reference missing.

Haptic Environment Identification with Friction
Stanford and Interval Corp

No results are given but the gist of the project is to determine the frictional properties of small devices and then be able to reproduce them through a haptic device.

reference missing.

Remote Palpation Instruments
Harvard
Robert Howe, Gregory Kovacs, and William Peine

This device extends their previous linear tactile display. An arm like device allows a tactile sensing apparatus to be inserted into the patient. The sensor relays the data back to the tactile reflection device at which point the data is portrayed to the user.

Peine, W.J., Kontarinis, D.A., and Howe, R.D., 1995, "A Tactile Sensing and Display System for Surgical Applications," Interactive Technology and the New Paradigm for Healthcare, Eds. R. Satava, K. Morgan, H. Sieburg, R. Mattheus, and J. Christensen, IOS Press, (Washington, D.C.), pp. 283-288.

Surgical Technology Research
Blake Hannaford, Mika Sinanan, Jacob Rosen, Jeff D. Brown, and Massimiliano Solazzo

Application of advanced technology to minimally invasive surgery training and procedures. They wish to investigate the biomechanics of these surgeries as well as improving current surgical instruments with force feedback technologies. They have put together a grasper with instrumentation as well as a force reflective grasper. Their instrumented version allows them to track forces and torques applied by a surgeon during an operation. This information is both recorded in real-time to a pc as well as composited onto the video display produced by the device for a picture-in-picture display. Their feedback device has resolution of 1400 quadrature over the full 0.6 radian motion range. The actuators come from the flat coil actuators of hard disk drives.

Hannaford B., J. Trujillo, M. Sinanan, M. Moreyra, J. Rosen, J. Brown, R. Lueschke, M. MacFarlane, 1998, "Computerized Endoscopic Surgical Grasper", Studies in Health Technology and Informatics - Medicine Meets Virtual Reality, Vol. 50, pp. 265-271.

Haptic Feedback for Surgical Simulation
University of Hull
GR Brookes and Derek Wills

The development of haptic feedback technology for surgical simulation is the main goal. They dont seem to be interested in the design of new tools, but are mostly interested in adapting current tools to be used by a haptic simulator. They have a 2dof device that is "suitable" for simulation and have also developed a 3dof version that allows translation along the tool axis. They state the current goal to be research into collision detection and force response methods able to support the high update rates.

reference missing.

Virtual Environment Knee Arthroscopy Training System
University of Hull
Derek Wills

A system was mae that allowed surgeons to practive arthroscoptic knee surgery. The made mock-ups of two surgical devices (the arthroscope camera and surgical probe) that the user would use in conjunction with a hollow articulated knee model. The position of the two devices is tracked and the graphical environment generates the appropriate images. The images produced are distorted to match those produeced by the barrel distortion of the true device. Collision detection was done using voxels. They also allow some interaction with deformable models (such as ligaments and menisci) by modeling the behavior using linear tetrahedral finite element methods. This limits their model complexity. They also are trying modal analysis which is less computationally complex.

Wills, D.P.M., Viant, W.J.V., Mohsen, A.M.M.A., Avis, N.J., Logan, I.P., Philips, R., "The Application of Technology to Orthopaedic Surgery," in Proc. of Interactive Technology in Surgery and Medicine Conference, British Computer Society, (Leeds, UK).

Eye Surgery Simulator
Georgia Tech
Mark Clements, Andrew Quay, and Ed Price

This system, and device, are patented and in use. They gathered both a collection of images from an eye bank, and haptic technology to accuratly simulate eye surgery. Apparently, they have the ability to do more than just simulate what the user is doing, but can alter the scenario (much like flight simulators do) in order to greater test the user. Ideally, resulting in better surgeons.

reference missing.

Overview

Introduction: Tactile
Introduction: Psychophysics
Introduction: Teleoperation
Haptic Cooperative Work in VE
Haptic (Sensory/Touch) Interfaces
Iwata Lab. Univ.of.Tsukuba Home Page

Haptic Devices

Images of devices

Haptic Rendering

Display Devices

Garnet Vision Display

Software Engineering

Pneumatic Haptic Interface (PHI) System
Colorado Haptic Interface
Iwata LHX

Networking

Virtual Coupling Networks