I.A.2. Title of Proposal:

Augmented Stimulation for Bodyworn Haptics   

 Thomas E. von Wiegand

 

I.B. Capable Manpower FNC Information

I.B.1. The proposed research involves the development of an electrocutaneous stimulator module for incorporation into a bodyworn haptics system being developed elsewhere as part of the ONR/VIRTE Demo II Close Quarters Battle (CQB) Trainer project. In discussion with the Demo II system designer and domain experts under contract to the designer, a need has been expressed for a method of producing stimuli, controlled by a virtual environment system, having a level of jarring agitation similar to that commonly experienced in real-world training. An essential function of CQB training is to introduce or maintain the user's acclimation to the level of apprehension engendered by actual battle, in which there are visceral consequences to one's actions. Absent such consequences, VE trainers for infantry-type scenarios may only achieve levels of immersion similar to that of interesting video games. Although we know of no pre-existing Navy / USMC requirement document outlining this need, it is commonly expressed as being an important component of effective training for this type of task.

 

I.B.2. Systems for bodyworn haptic display are typically designed to serve as either spatial displays of data (e.g., visual-haptic sensory substitution) or as alerting systems. Such displays have been designed in the past using direct electrocutaneous stimulation, but the recent trend in the area of VE research has been to utilize arrays of electromechanical actuators. The present goal is to develop a safe and effective method of generating a high level of cutaneous stimulation exceeding that which is possible using vibrator or motor -based, stimulators. The technical approach to be used involves the generation of controlled cutaneous surface currents, preferably induced without direct skin contact. This approach is more straightforward relative to D.C. and low-frequency methods in which galvanic contact is made, and thus will be the primary approach to be explored. The end products will consist of design specifications and a sufficient quantity of prototype units to be incorporated into a bodyworn haptics system. Given prevailing theories concerning the desirability of heightened emotional states as elicited via realistic stimulus presentation, we expect the payoff for the VIRTE program to be greatly enhanced user immersion in the task and improved training effectiveness for stressful combat situations.

 

I.C. Technical Information

I.C.1 Requirement / Problem / Deficiency

The operational deficiency stems from the adage that "soldiers will fight the way they have trained." Real training for infantry tasks (such as CQB) involves a certain amount of bodily contact, often to the point of pain. Heretofore, VE interfaces have been designed primarily as portals to relatively physically-neutral worlds. Some work has been done on introducing fear-inducing physical situations using visual or haptic tricks, however, the range of stimuli conveyed to the user is limited to the dynamic range of the output device, which in almost every case is feeble enough to be "tuned out" by a jaded user. The cost of ignoring this deficiency within a new training modality such as VE could be grave. Systems that utilize existing "low impact" actuators can undoubtedly accommodate the informative aspects of training and may even accommodate the goal of avoiding "the instillation of bad habits that may resurface in combat." However, if training has not been sufficiently stressful, then soldiers will react to a real combat situation in unreliable ways (indecision, forgetting known response drills, lack of motor coordination, etc.). Realistic training that incorporates the adrenaline rush of imminent danger can alleviate this shortcoming of VE-based systems. S&T is required to develop devices that can safely address this particular requirement, in a practical VE training system. Application of robust multipoint mechanically-grounded haptic stimulation is not a practical option within the next decade, and the forces that can be generated using non-grounded bodyworn mechanical arrays (such as vibrators) are relatively limited. Within a variety of domains, electrical stimulation has been applied in situations ranging from mild (sensory substitution displays for speech and hearing) to violent (bark control, cattle prod, electric fence) stimulation of human and animal tissue. A great deal of literature exists on electrocutaneous stimulation, and a well-developed technological base (of similar, primarily medical, products such as TENS devices, Faradic stimulators, muscle electrotherapeutic devices) is available which can be adapted to the particular needs of the VIRTE Demo II system. Without a specific S&T effort it is unlikely that a device having the particular requirements needed for military training (notably, density of stimulation array and dynamic range of stimulation) will be produced in the commercial (security, medical, or entertainment) arena. We know of no existing MNS/ORD documents relating to this novel requirement, and to our knowledge no policy or doctrine changes should be necessary.

 

I.C.2 Technical Background

We have begun related work in the area of bodyworn haptics for application to VIRTE Demo II in our previous grant "Realism in Virtual Environments." Other relevant VIRTE research on bodyworn systems is being done by Robert Lindeman's group at GWU. This work utilizes vibrotactile elements and would be readily enhanced through incorporation of the electrocutaneous stimulator modules being proposed in this document. We hope that our stimulator modules will contribute to the ongoing work of other groups working concurrently on other aspects of the bodyworn haptics system for VIRTE Demo II.

As stated in the previous section (I.C.1.), there is a long history of research in the area of electrocutaneous stimulation, much of which has evolved into commercial products that share some of the characteristics of the proposed devices. The basic electronic and biological aspects of the work are documented in existing literature, but the specific engineering issues involved in producing a stimulation system that encompasses the desired dynamic range, distribution of stimulation, degree of safety, ergonomics, and power requirements, all in a self-contained module of small physical volume, have not been previously addressed and hence are the subject of this proposal. (See the References section for an abridged list of this background literature.)

 

I.C.3 Technical Approach

1. Overview

Representation of virtual contact is an important component of any realistic CQB training system. The associated input to the trainee can be achieved using direct approaches based on mechanically-grounded robotic arms or indirect, but more practical, approaches based on sensory substitution (either within or across modalities). Independent of how the virtual contact is achieved, such contact can be used not only to signal collisions with virtual objects, but also to signify more abstract conditions or events directly related to the goals of training. The primary purpose of the work proposed here is to develop a tactile stimulator module that can be incorporated into a bodyworn haptic array. Such an array may be used in the training system to signify (1) exposure to enemy fire and (2) the impact of being hit by a round or other projectile. We believe that the signaling of such conditions and events constitutes an extremely important function of bodyworn haptics in CQB training, and that the proposed module is ideally suited to this application.

Work on the module can be broken down into the general stages of stimulator development (research and design), and prototype implementation/testing (i.e., construction of prototype modules to be supplied to the systems integrator, and subsequent communication with end users regarding module effectiveness). Later in the project (part of phase II) we will explore the benefits of integrating the individual modules into a cohesive substrate in which construction takes advantage of "E-Textile" technology. The expected advantages in ergonomics as well as technical performance are likely to warrant this additional development effort.

 

2. Stimulation System Background

Safety, power, and ergonomics are integral considerations in the design of the stimulator units. The envisioned unit will operate in two signaling regimes: (1) a mild signaling of exposure and incidental contact; and (2) a strong signal of impact, such as being shot. Each of these two regimes can be effected through electrical stimulation using either capacitive or direct (galvanic) coupling. The development work involves practical research exploring each signal regime and coupling technique, and determination of how the methods stand up with respect to the major criteria of ergonomics (and user acceptance), interference with ancillary electronic equipment, as well as effectiveness within deployable training systems.

We believe that direct electrical stimulation has the potential for becoming the preferred signaling modality in bodyworn haptics systems. The unique sensible qualities of electrical stimulation (due in part to the unusual characteristic that the psychophysical power-law exponent is much greater than 1) gives the sensation a unique urgency. Also, electrical stimulation devices can potentially be made lighter and more compact than other stimulators, and are more amenable to replication in dense arrays as part of e-textiles. There is a long history and extensive literature on electrical stimulation techniques (e.g., Tousey 1910); more recent applications include skin surface arrays for auditory and visual substitution (Saunders 1974, Sparks 1978, Cowain 1992, Kaczmarek 1992) and arrays designed to be placed in contact with the mucosal surfaces of the mouth (Kaczmarek 2001). These applications attest to the controllability, safety, and feasibility of the approach.

The present goal is to achieve stimulation in both regimes (mild exposure and sharp impact). While it is relatively easy to produce unpleasant shocks to satisfy the latter case, producing mild sensations involves a more measured application of current based on actual skin conditions. Controlling the skin conditions both actively (through the use of skin conductance measurement and automatic control of the stimulation parameters) and passively (through innovative electrode design) are two methods of achieving the appropriate degree of stimulation. Typically, the passive method involves the use of messy electrode gels or creams in order to control the skin-to-electrode resistance. Clearly, this approach will not be acceptable to users of the system. To address this ergonomic issue, we have identified a number of commercially-available electrode designs utilizing conductive plastics and/or conductive (non-soiling) adhesives. Part of the work proposed will be to test various electrode types for effectiveness and user acceptance.

An area of great promise is the use of high-frequency electric discharge, which allows the signal to travel capacitively (through clothing) without the need for galvanic contact. The major technical issues involved will be reducing the size of the components typically used to generate such signals (transformers/inductors) and ensuring that any radiated signals do not interfere with other nearby electronics. Although this method has been less popular as a means for achieving sensory substitution, it exhibits intrinsic safety features in that there is (a) no direct connection of the apparatus to the skin (avoiding unexpected leakage paths) and (b) the high-frequencies used tend to constrain currents to the surface of the body (thus avoiding any possibility of current flow through vital organs).

 

3. Biological and Electrical considerations

There are multiple mechanisms by which the coupling of electrical energy to the skin results in a perceptible sensation. Due to the electrochemical basis of neural action, direct stimulation of neurons via exogenous electrical currents is the most obvious route to sensation. A complicating factor is that the neural "circuitry" is in large part insulated from electrical excitation by the stratum corneum (at the skin surface) and by fatty processes such as the myelination surrounding the nerves. Appropriate selection of electrodes, and the applied waveforms and frequencies, allow these factors to be overcome to a certain extent. Also, advantage can be taken of the non-homogeneous nature of the skin structure: for example, the sweat ducts introduce electrical shunt paths due to the high conductivity of the channeled saline content. A second mechanism of perceptible sensation is due to direct electrical action on the muscles or motor nerves. The degree of stimulation extends from barely-noticeable, through sub-tetanizing stimulation (fairly common in therapeutic settings), up to full tetanic convulsing of muscle groups (ranging from dramatic to painful). A final mechanism of sensible stimulation involves the heating effects of the energy flowing into the tissues. The mechanism of sensation in this case is the same as for any conventional source of heat and is of little interest given the inefficiencies of electrically heating tissues (water has a high specific-heat value). However, transient heating occurring over small areas often accompanies punctate discharge and can serve to augment the other mechanisms.

The effects of the transfer of electrical energy to the skin depend on a variety of well-understood (or at least well-characterized) factors. Of primary interest in the current development work is the psychophysical relationship between sensation and the physical stimulus applied, often described as an exponential function (power law of sensation, after the work of S.S. Stevens). Whereas the exponents for "distal" senses tend to be around 1/2, (sensation grows roughly as the square root of stimulus intensity), and exponents for most "proximal" senses are around 1 or slightly higher, the exponent typically reported for electric shock is 3.5 or above. This means that the sensation grows much more rapidly than the physical stimulus. The consequence of this relationship is that electrical stimulation is a very efficient means of generating strong (or even painful) sensations. The selection of stimulus frequency and waveform determine the spatial distribution and physiologic action of the energy transferred. These effects are accounted for by modeling the reactive and geometric aspects of the electrode and tissue, as well as modeling the responsivity of the nervous system to various periodicities of stimulation.

 

Heating is another effect of energy transfer to the skin. For the present purposes it is considered as a limiting factor due to the long thermal time-constants, and in the case of large amounts of energy transfer, the danger of thermal damage to tissue. In general, the heating effect is proportional to the square of the current density at the electrode (i.e., the temperature rise in a tissue volume is proportional to the square of the current density). In the proposed device, all current flows will be limited to levels below which gross thermal effects are significant. Another effect related to heat generation arises in the case of extended electrical discharges (arcing). In addition to the damaging heat generated, the significant actinic power of extended arcing results in chemical action on the skin (ultraviolet burns) and reaction with air (ozone production). In the current work these conditions will be avoided through power limiting.

Under certain conditions, such as prolonged application of D.C. currents, chemical and biological changes at the tissue interface can occur. Such effects are used for example in electrolysis (depletion or accumulation of ions), electrophoresis (protein and cell migration), iontophoresis (ion migration), electro-osmosis (volume transport). Although these effects are useful in certain clinical situations, for safety reasons we wish to avoid them. Accordingly, the stimulus waveforms to be used in the present application will have no D.C. component (i.e. will be D.C. balanced).

Methods of coupling electrical energy to the skin must take into account safety, convenience, and effectiveness. Starting with an appropriately simplified model of the geometric and electrical properties of the skin, one can design electrode configurations to accommodate the preferred excitation mechanism, electrode carrier or substrate, as well as other design requirements. For example, an important consideration is whether to utilize high frequency stimulus waveforms which are compatible with capacitive coupling. Such electrodes could be more convenient since they do not need to make direct contact with the skin and could be applied over clothing. The drawback to this approach is that the capacitive electrodes act essentially as antennas, and as such may cause interference to neighboring electronics. Additionally, it can be difficult to control the spatial relationship between the electrode and the skin surface if applied over loose clothing. Direct (galvanic) coupling has the advantage of a stable geometry with respect to the skin surface. However, galvanic coupling obviously requires direct access to the skin, and also requires special attention to the avoidance of unintended current paths between unrelated electrodes (and external circuits). A number of approaches are available to minimize these shortcomings, among the most promising that we plan to exploit in this project are: multi-conductor electrodes constructed to precisely control current paths, special conductive polymers to ease connection and removal from skin surface, and electronic-textiles containing conductive paths allowing the user's (admittedly specialized) underwear to serve as an electrode array.

 

4. Design Considerations

A summary of the design considerations under which development of phase I and phase II prototypes will proceed includes five main elements:

* Safety

* Adequate dynamic range

* Ability to deploy multiple units in extended array on the body

* Non-encumbering, easy to use, put-on & remove

* VE system friendliness

 

Safety of the user is of paramount importance if the module is to become part of a training system. Well-documented design practices will be followed which will result in a safe system, including but not limited to: isolation of power supplies, control signals, current loops (through body), and limits on energy flow and duration. The concept of electrical stimulation can also be somewhat frightening to the uninitiated, given common dramatic depictions of ECT (electro-convulsive therapy), capital punishment (via electrocution), etc. Although a certain amount of fear may be useful for the training purposes of the system, special care will have to be taken in how the system is presented to the user in order to avoid unwarranted duress. The goal is to create a device that is actually safer than devices that are currently used in "real world" training, such as paintball guns.

Adequate dynamic range refers to the ability of the device to produce sufficiently powerful stimuli, while also accommodating the softest desired signal. Since mild stimuli can be produced using other more conventional methods (such as vibrators and pager motors), it is vital that the proposed device easily produce stimuli at the more intense range, even if this means some loss of ability to reliably produce soft signals.

Ability to deploy multiple units is vital if the system is to be used for general indication of contact anywhere on the body. Although saltatory interpolation has been demonstrated for haptic stimuli, it is a clear requirement to involve as much of the body surface as possible.

Non-encumbering implies that individual units must be small in volume, light in weight, and take advantage of favorable electrode configurations and substrate choices.

VE-system friendliness is an important consideration since the rationale for the stimulator is that it facilitates a VE training system. If the stimulator causes interference or malfunctions of other system components it is of little practical use. Special considerations of signal emissions as well as ease of interfacing to the bodyworn haptics circuitry it is to be a part of are important design issues.

 

5. Development Tasks

Creation of the stimulator module will proceed in a fairly sequential manner, part of which will be iteratively repeated. Before a repetitive design cycle can begin, we will develop initial computational models representing skin and electrodes. Calculations of energy, waveform and electrode design will be based on the results of an extended literature search and interaction with results of the computer models. A testing device will be built in order to quantify and compare the effects of various electrode/waveform configurations, and subcomponents of the module will be breadboarded in a modular fashion to facilitate the testing of design changes.

The formal production cycle will include design of the major components of the module (energy conversion/signal generation, triggering and power control, power supply and energy management). The components will be integrated and evaluated for desired functionality. Once the design is finalized, the phase I "production run" will begin: printed circuit boards will be drafted and produced, parts will be ordered, the phase I prototype modules will be constructed and tested (and repaired as needed). At this time, the documentation will be refined so that the modules can be delivered and integrated into the rest of the bodyworn haptics hardware.

We will obtain user evaluation and feedback, according to guidelines that we will suggest. This feedback, plus further design efforts on array production and electrode design (utilizing conventional as well as novel technologies such as electronic textiles), will provide the basis for the next iteration of the design.

The phase II production will proceed similarly to phase I. The evaluation after delivery of the phase II prototypes will be much more elaborate, and will involve extensive consultation with domain experts. In addition to possible device refinements, the main goal of this consultation will be to develop guidelines for the application of the new technology to training systems. Both technical considerations (integration issues) as well as curriculum considerations will be addressed at the end of the second year of the project.

 

 

I.C.4 Dual-Use

Research in this area is specifically tailored to VE training of infantry tasks, though extra-military use is possible. We have not identified any industrial partners, but the work could be of interest to commercial firms involved in security and animal control. A prototype for a related type of stimulator, called the "Bioforce Game Controller" has been shown at gaming conventions by Mad Catz Inc., demonstrating the interest that video game manufacturers may have in the more powerful and extended system to be developed through the proposed research.

 

 

 

 

References

 

Cowain, R.S.C., P.J. Blamey, J.I. Alcantara, P.A. Blombery, I.J. Hopkins, L.A. Whitford, G.M. C.ark, 1992. "Safety studies with the University of Melbourne multichannel electrotactile speech processor," J.Rehab. Resch. & Dev. 29:1, pp 35-52.

 

Enderle, J.D., S.M. Blanchard, and J.D. Bronzino, 2000. Introduction to Biomedical Engineering. Academic Press, London.

 

Gregory, R.V., W.C. Kimbrell, and H.H. Kuhn, 1989. "Conductive Textiles," Synthetic Metals, 28, No. 1,2, C823-C835 (January).

 

Grimnes, S., and O.G. Martinsen, 2000. Bioimpedance and Bioelectricity Basics. Academic Press, London.

 

Kaczmarek, K.A., J.G. Webster, and R.G. Radwin, 1992. "Maximal dynamic range electrotactile stimulation waveforms," IEEE Trans. Biomed. Eng., 39, pp 701-715.

 

Kaczmarek, K.A. and M.E. Tyler, 2000. "Effect of electrode geometry and intensity control method on comfort of electrotactile stimulation on the tongue," Proc. ASME Dyn. Sys. Contr. Div., Orlando, Florida, pp 1239-1243.

 

Lind, E.J., R. Eisler, G. Burghart, S. Jayaraman, S. Park, R. Rajamanickam, and T. McKee, 1997. "A Sensate Liner for Personnel Monitoring Applications," Proceedings of First International Symposium on Wearable Computers, Cambridge, MA, IEEE Computer Society, Los Alamitos, CA, pp. 98-105.

 

Saunders, F.A., 1974. "Electrocutaneous displays," in Cutaneous Communication Systems and Devices, edited by F.A. Geldard (Psychonomic Society, Austin, TX), pp 20-26.

 

Sparks, D.W., et al., 1978. "Investigating the MESA (multipoint electrotactile speech aid): The transmission of segmental features of speech," J. Acoust. Soc. Am. 63, pp 246-257.

 

Reilly, J.P., 1992. Electrical Stimulation and Electropathology. Cambridge University Press, New York.

 

Tousey, S., 1910. Medical Electricity and Roentgen Rays. W.B. Saunders Company, Philadelphia.

 

 



Augmented Stimulation for Bodyworn Haptics Stim.htm @ 12:59 PM, July 13, 2001