Wearable Computers

CIS-732

Fall 1999 Semester

Professor Turoff


Copyright © 12/16/99

Brian D. Rauch

available for free educational use with prior permission of author

email the author:  mjbrauch@uscom.com

Submitted to professor by Brian D. Rauch: 12/9/99

 

Table Of Contents:

Chapter
Subject:
Page:
1.0
Introduction
2
2.0
Background
2
2.1
Xybernaut: A Current Commercially Available Wearable System
3
3.0
Networking and Operating System
4
3.1
Personal Area Network
4
3.2
Pointing System
5
3.3
Wearable Computer Component Pieces
5
4.0
Agents
7
4.1
Recognizing Affect
8
5.0
Applications
9
5.1
Smart Wallet
9
5.2
Worker Training and Worker Support
9
5.3
Smart Houses
10
5.4
Smart Eye Glasses
10
5.5
Smart Cars
11
5.6
Military
11
6.0
Handicapped
11
7.0
Ergonomics
12
8.0
Future Research
13
Bibliography
15

1.0 Introduction

Wearable computers are described as the next wave of technological innovation and breakthrough. The computer evolution has moved from mainframes, to the desktop, and now the computing power is moving onto the person. The term "wearable computer" was coined by a research group at Carnegie Mellon University in 1991 (Siewiorek, 1999). By the user actually "wearing" the computer, s/he can utilize the power and functionality virtually anywhere in their environment. Moving the CRT screen off the desktop and into a miniaturized "eye-piece" enables the user to carry on activities (e.g. - face to face conversation) while never looking away from their computer. This technology is developing and changing rapidly primarily due to vast changes in computer speed, wireless technology and miniaturization of components. Wearable computers have extended into many diverse domains creating development problems in the integration of software and hardware as designers respond to the many application challenges (Smailagic, 1997).

Mann (1997a) defines wearable computers as having the following criteria:

  1. the physical computer hardware is seen by the user as a somewhat natural extension of themselves. In this manner, the computer is not connected to a desktop nor any electrical outlets to hinder the user, the user is to feel free and unencumbered by the computer
  2. the user is always in control of the computer, even with minimal conscious effort the user controls computational activity, such the user sees the computer as a cognitive extension of themselves.
  3. the computer is fast, so processing time is minimal. The computer is always active while being worn, even though it may be a 'sleep' mode, it can awaken (on its own) given the correct environmental cues. Lastly, the user is always in connection to the computer via one or more output methods (eg. - video display)


2.0 Background

Bradley Rhodes (1998?) outlines the growth and development of the field of wearable computers starting with the invention of eyeglasses in 1268 (addition of an artificial device to augment human sense organs), leading to the 1997 IEEE International Symposium on Wearable Computers in 1997. Between these two events he lists many milestones such as Nathan Kline defining the word "Cyborg" (1960), Thorp and Shannon's development of a miniature computer system to track the rotations of a roulette table and sending a signal to the user via a hearing aid (1966), and the first wearable computer head mounted CRT display (1966). Other significant developments included using a radio link (1972), including a camera mounted unit (1977), the wiring of a wearable "shoe" computer which was also used to beat the odds on a casino roulette table (1978), the introduction of the Sony Walkman (1979), Steve Mann's first computer "photographer's assistant" using an Apple-II mounted into a steel-framed backpack (1981), commercially available head mounted monochromatic video display with a 1.25" diagonal screen which appeared to be 15" with display screen worn close to the eye (1989), the development of a name badge using infrared signals to communicate the location of a person within a building (beginning of "smart rooms") (1991), global positioning is added to available functions (1993), the intelligent agent (Remembrance Agent) by Thad Starner is used (1993), introduction of the wrist computer and belt computer (1994), Wearable Computer conferences by DARPA and Boeing (1996) and a "smart clothing" fashion show held in Paris and Boston (1997).

With advances in technology the component apparatus of wearable computers are becoming smaller, lighter and much more efficient. Thus the prospects for rapid growth into more mainstream markets is strong. Such was not the case in early prototypes however. Steve Mann describes his early experiences:

"…My wearable computer efforts of the 1970's were a success in some ways, and a failure in others. The success was in demonstrating the functionality of an early prototype of wearable computer system, and in formulating a practical application domain for wearable computing. However, there were many technical failures. In particular, the bulky nature of the apparatus rendered it often more of a photographer's burden than a photographer's assistant. Furthermore, the reliability problems associated with so many different system components were particularly troublesome, given the nature of typical usage patterns: walking around on rough terrain where wiring and the like would be shaken or pulled loose. Interconnections between components were found to be one of the major hurdles" (Mann, 1997).

Wearable computers provide numerous applications, and are especially helpful in "hands free" tasks like maintenance operations. One such prototype was developed for use of repair of large vehicles such as trucks or aircraft (Bass et al., 1995). These types of complicated repair tasks often call for sophisticated knowledge and computer assisted testing/trouble shooting. Additionally, the repair person may be working in very limited physical space, so being able to refer dynamically to a computer (referring to the repair manual) while simultaneously keeping both hands free is a great asset.

2.1 Xybernaut: A Current Commercially Available Wearable System

Sony corporation has developed and is currently marketing the Mobile Assistant IV (MA IV), or the Xybernaut system. This is a speech activated system using a head mounted display and is targeted for workers in the airline, manufacturing and medical industries (Wired Magazine, 5/15/98). The system is based upon Intel's "Tillamock" chip using a 233 MMX mobile processor, with 128 MB RAM, 4.3 GB hard drive and uses either Win95, Win98 or WinNT operating system. The price is $6,971.00 for the full package (Lithium-ion battery with AC charger adapter, 11" mini keyboard, belt, mini part replicator for floppy drive and keyboard, operating system and a head mounted video Xybercam) (http://www.xybernaut.com/). The MA IV does appear to be the most functional prototype wearable computer system to date (Ditlea, 1999). The Xybernaut website, http://www.xybernaut.com/, lists various functions of which the wearable package is capable of:


3.0 Networking and Operating System

The goal of wearable/interactive computers is to network all components through one operating system. For example, a smart house would have a network connecting the toaster, curtains and phone. Hughes (1998) reports that there are two competing operating systems, Sun Microsystems' Jini and Microsoft Windows CE ("Compact Edition"). Sun apparently sees all household appliances/components networked through Java/Jini interface, while MS seems to be taking the more traditional personal computer interface approach in connecting appliances to the network. The Linux 2.0 OS is being used by Steve Mann for his WearComp/WearCam wearable computer network (Mann, 1997a).

Wearable computers are increasingly needed to provide a larger scope and variety of tasks, thus being even more versatile than some desktop systems (and as such necessitating more system resources than are available. This has been a limitation of wearable computers, in terms of their restrictions in main memory, secondary memory, processing speed and power. Given a large variety of tasks, it would be impossible to pre-load all software which would be potentially needed upon the wearable computer. Fickas et al. (1997) developed a software architecture "middlelayer" where the limited internal resources are managed by a "decision maker". This decision maker monitors the current system status and decides, for example, whether a task should be run locally, or remotely on the networked server (via wireless connection).

3.1 Personal Area Network

As the number of wearable computer peripherals increases, we are faced with an increasing amount of devices which need to be linked (such as cellular phones, personal digital assistants, pagers, etc.) hearkening in the era of the "Personal Area Network" (PAN) (Zimmerman, 1996, Post et al., 1997). With all the variety of devices composing a wearable computer system, there will be input and output duplication with multiple keyboards, microphones and displays. With limited (or no) communication between these many diverse devices this opens up a need to find a method to share I/O, storage and computational resources. The PAN would also be connected to other networks. Zimmerman (1996) describes a scenario:

"…I am at home preparing for the day and want to find the time of my first meeting. I call out 'When is my first meeting?' The microphone in my watch transmits my voice through a series of transponders distributed through my house to a voice recognition computer that searches my calendar and sends back a response to a speaker or visual display in my watch. When I leave my house the door senses my departure and sends a message to my colleagues. When I approach my office building, the door acknowledges ne by opening, sends a message of my arrival to my colleagues, and uploads any new messages" (page 609).

In terms of how the PAN connects components across the wearer's body Zimmerman speaks of the difference between 'near-field' and 'far-field' communications. Far-field (e.g. - radio) is more susceptible to interception, interference and eavesdropping, and is also subject to government regulation. Near-field communication is much more secure (thus insuring privacy of personal information such as credit cards and cell phone numbers), uses less energy and operates at a much lower frequency (0.1 to 1 megahertz). Post et al. describes near-field communication in terms of how the "electrostatic field strength decays as the cube distance from the source" (1997). Zimmerman talks of using the human body as a "biological conductor" for near-field communication which connects all pieces of the PAN.

Post et al. has expanded upon the above work and has been able to create a wireless transmission of electrical power (1997).

3.2 Pointing System

For wearable computers, the finger can be used as a pointing system ("finger mouse") in conjunction with a head mounted camera (Starner, 1997, Mann, 1997b). The camera view is seen in the eye piece, and superimposed on the view in the eyepiece is the menu. The user moves his/her finger in front of the glasses - this is captured on the camera, and the relative position of the finger vs. the see through menu chooses which option the user desires. Another input device was a large dial with a selection of buttons (Bass et al, 1997). The dial is analogous to an early radio dial and one advantage is that it can be used with gloves making it versatile for many applications. Yet a third input system is a miniaturized keyboard strapped onto the forearm. An example of this wrist keyboard is a half-QWERTY keyboard (Matias, 1994) which contains all of the keys on a standard QWERTY keyboard (one side only), with the keys of the other side absent, but with the user being able to access these other keys via the use of the spacebar, thereby only needing to use one hand. A last form of data entry is using speech recognition to issue commands to the wearable computer.

Another challenge is how the network will be physically connected, via wires, or wireless. The capacity for either exists with current technology, however, for pricing purposes, wireless is more expensive at present.

3.3 Wearable Computer Component Pieces

  1. Camera - head mounted cameras can provide the user with different view (zoom in/out), as well as provide a mechanism for accessing menu choices (through a combination of visual display and hand motions - see Networking and Operating Systems). Additionally, the camera can be linked to a processing application whereby images can be interpreted. For example, by conversing with another person, body language can be programmed so as to be read and analyzed, or specific objects in the environment can be identified with associated messages and prompts being provided to the user (Starner, 1997). Additionally, a camera system can be used in conjunction with some electronic identification system in giving tours, when the user entered, let us say, a specific room within a museum, information on artifacts with in this room get downloaded on to the computer. The computer is programmed to recognize certain physical objects, and when these show up within the camera lens, the computer will high-light these and the user only has to indicate that more information is requested on this object and this information will be provided. In this manner, the user is in control, and can explore at their own pace and leisure, at the same time, the camera processing continually recognized and tracks the objects that the user comes into contact with. This could also be used to interpret sign language and even remember faces. In the case of the latter, when a face has been successfully recognized, then specific information including name and background data can be supplied to the user. Most current technology has the camera as part of a pair of what appear to be ordinary glasses, or sunglasses (Mann, 1997, Lieberman, 1999).

  1. Positioning System - a mechanism where the user of the wearable computer can both track their physical location, or be tracked by others. One method is using the Global Positioning System (for outdoor use), another method is setting up a system of infrared light powered beacons (for indoor use) (Starner, 1997). By having the wearable computer communicate with these infrared transmitters, a fixed position can be maintained, and remembered for future reference. In this manner, it would be possible for the first user to place a memo-message in the virtual visual field which, when the second user passes by this same sensor, the message will be displayed. These messages, or 'tags' can be useful in repair of machinery by the manufacturer to aid in maintenance and for added guidance. In this manner, specific instructions can be displayed in the user's visual field, thereby providing graphic and text displays in a context sensitive environment.
  2. Battery Power - there is a need for power for components, although a small solar collector can assist to a small degree, batteries are still necessary.
  3. One Handed Keyboard
  4. PC-104 based 50 MHz 486 computer, 16M RAM, 1G hard disk (Starner, 1997); but new operating systems now incorporate Pentium 233 with 64 M RAM and 4.3 GB (http://www.Xybernaut.com). A comparison of these two hardware packages, roughly two years apart is an indication of the growth of this field and a harbinger for future growth and development!
  5. Bio-Sensors (used in "affective computing")
  6. sound board and video board
  7. CRT or LCD display
  8. extra disk capacity
  9. custom clothing
  10. Monitor: the early video displays were small boxy monitors which attached to a head strap and fitted entirely over one eye. These have developed significantly and the current state of the art monitor is using what seems to be a regular pair on sunglasses, upon one lens in a print out of data across the inside of the lens (these sunglasses also contain a camera). Mann speaks of "homographic modeling" where the user can place a 'virtual post-it note' in one location, and this note can be later viewed (within the video monitor sunglasses) by someone else returning to this same location (1997). Additionally, this note can be reserved for a specific individual, thus the note remains dormant until intended receiving user approaches the scene, then the note is activated within the new user's video monitor.
  11. Input device: using a dial (Bass et al, 1997), finger mouse (Starner, 1997, Mann, 1997b) or a hand-typing wrist device (Matias, 1994).


4.0 Agents

A premise of a wearable computer is that the computer can learn and become customized to the user's needs, based upon the task the user is involved with. Starner (1997) and Bradley (1997) speak of a Remembrance Agent (RA) which provides context sensitive suggestions of information and files. For example, if the user is viewing a document, the RA may suggest an associated email, showing minimal information such as the subject, sender, date, or what ever information the user would want to view to get a sense as to decide whether the entire email should be retrieved. Similarly, if the user is having a conversation, the RA will suggest related data topics in the user's personal database (through use of voice recognition software via a wearable microphone). Humans are poor at data retrieval, and a complex hierarchical directory quickly breaks down when the human user loses the thread of where the item being searched for resides. When using a desk computer, the computer cannot be of much assistance because the context of the use of the computer is always the same (human working at desk inputting data). However, within the rich environment of the user's world the potential for useful contexts increases significantly. A computer remembrance agent can take cues from this environment to assist in locating appropriate files from memory, picking up on cues like where the user is, is the user alone or with someone (who is the 'someone' the user is with), what is the user doing at that time, etc., all can be used to search and locate appropriate files that may be relevant to the user at that time (Bradley, 1997). In this manner the RA is a continual presence, always 'looking over the shoulder' of the user. For example, at a staff meeting the RA can assist in remembering names of attendees, who was at the meeting prior, what they said at that time, what was on the agenda, and also to suggest topics which might be of relevance to what was going. The reference database from which the RA will pull information can be the user's own file structures, or could be some other source (network, Web). The RA only provides one line summaries of files, emails, notes, etc, at the bottom of the eye level mounted display where it only takes the user one command to retrieve the full file.

The RA can also receive affective user data through body sensors (Picard, 1997). By being aware of the emotional state of the user, another layer of information is added to other data (e.g. - environmental), all enhancing the quality of interactively between the RA with the user. The RA is always active in searching the user's own personal data base for information which might prove useful within the context of the current task the user is performing. Using affective body sensors, the computer can sense when the user is preoccupied or dozing (during a meeting or class), and then begin a recording of the missed activity so the user can refer back to review what was missed through inattention. Through the use of a wearable computer, the "intelligent agent" can be a consistent and reliable presence to assist the user in real time tasks that are not possible while the user is tethered to a desktop system.

In terms of using an intelligent agent to assist in performing manual work tasks like factory work and complicated repair, the "intelligent monitor" can anticipate what worker problems are likely to come up and then automatically link the worker to the component of the associated database system most likely to assist the worker in that situation (Najjar et al., 1999).



4.1 Recognizing Affect

Our emotions play a significant role in our every day life, and up to now, this had been unrepresented within human computer interfaces where the only connection between the user and the computer has been finger on keys or fingers on the mouse. As alluded to above, the wearable computer offers real potential to tap into the ambient user mood state (Starner, 1997, Picard 1997). Human vital signs such as blood pressure, temperature, galvanic skin response, foot pressure and electromyograms can be tracked and monitored through the use of body sensors connected into the wearable computer. In this manner, the computer becomes physically connected to the user and can learn to recognize physiological states of being which can then be related to a discrete feeling, or affective state (fear, stress, being relaxed, happy, or engaged in a task). By this monitoring of body vital signs, and learning how the user reacts when in different feeling and mood states, the computer can take action accordingly. For example, if the user likes to hear a certain type of music at one time, and another during a different mood state, the computer can plan on providing this to the user at the appropriate times. In this manner, the computer is predicting what the user would prefer based upon the many divergent data coming in through the camera (environmental setting) as well as the emotional data received through the bio-sensors.

Affective computing in wearable computing can also tell us more about ourselves. Picard & Healy note that the overall understanding of what emotions are, from a psycho-physiological perspective is not entirely clear. The same physiological reactions (e.g. - increased respiration, increased heart rate) might mean fear in one person, but mean something entirely different in another. In this manner, the data from affective computing could further help in the understanding of human emotions and provide a testing ground for emotional theories (Picard & Healy, 1997). The affective sensors could also be of assistance in medical monitoring where important health data might be tracked and reported directly to the user's physician. Of course, the user always has control over how the data gets collected, how the data will get used and, of course, what physiological data would be collected in the first place. They list four bio-sensors used in a current prototype affective wearable system: measuring respiration, skin conductivity, blood volume pressure and electromyogram, all of which the sensors can monitor painlessly from the surface of the skin. These four sensors are monitored via a Linux operating system, and can be sampled up to 20 times per second.

Again, as alluded in the above section (4.0), affective computing employs its own intelligent agent which can make decisions based upon analysis of the incoming environmental and physiological sense data. In measuring specific user input data, and user response data that the user is confronted with over time, the computer learns to act based upon what the user has desired in these situations in the past (or, the computer could extrapolate what worked, or didn't work, for the user in a similar situation). For example, as mentioned, when the affective intelligent agent determines that the user is under stress, it might play a certain piece of music which, in the past, has proven to decrease the physiological stress reactions upon previous playing. In this manner, the wearable computer system is adapting to the user.

The drawbacks to the current state of affective computing is the reliability of the bio-sensors and the need to have more efficient computer monitoring equipment which is power efficient, more accurate and more comfortable to wear. Additionally, there is a dearth of information regarding how to really understand bio-sensor data in terms of actual emotion state data they are conveying. Clearly this part of wearable computing is still in its infancy.



5.0 Applications

Wearable computers are developing applications in numerous areas too diverse to be included in this research overview. Cell phones, portable CD players, pagers, overnight shipping company bar code readers, etc., could all be referred to as wearable computers, but will not be explored here. With increasing computer power and miniaturization of components, horizons for wearables, like the digital revolution itself, are expanding by the hour. Below are a few commercial applications.

5.1 Smart Wallet

Jim Louderback recently reported on an electronic wallet (1999). The WEARLOGIC SmartWear Electronic Wallet advertises as a full computer contained in your wallet. This "smart wallet" contains a display screen, keypad and the user can input data such as telephone numbers. Additionally it can perform calculations, store data, and there is a special function which can store and display the bar codes on plastic cards (like library card bar code) so the user can actually carry less plastic in his/her wallet.

5.2 Worker Training and Worker Support

The use of wearable computers are being used to support factory workers who are challenged with downsized staffing and increasingly complex tasks to perform (Najjar et al, 1999, Siewiorek, 1999). Especially in the area of training an employee to perform specific duties the wearable computer offers significantly greater range than traditional techniques. Traditional learning takes place away from the actual environment where the learned task is to be performed, in prescribed times and expecting that the learner takes what has been learned in the classroom and then apply it to real-time job situations. Using a wearable computer allows the worker to be trained at the job location while actually performing the job task to be learned. The learner can be given immediate context sensitive feedback, progressing at their own pace, getting the specific feedback and assistance they need when they request it. Wearable systems also allow for closer tracking of quality, and allows for assessment of task completion. Additionally the software can be proactive in prompting, or intervening to suggest or bring attention to some detail to assist the worker in the task completion. This would not only lead to more efficient task completion but would also greatly increase safety and could diminish error rates. This was exactly the case when Georgia Tech Research Institute investigated the use of wearable computers in a food processing plant (Najjar et a., 1997). In this study the researchers built a wearable computer system and concluded:

The Factory Automation Support Technology (FAST) is an example of such a system. FAST is a database of information specific to the completion of a task. Included in this database would be on-line operating procedures, product information, specifications, repair manuals, policies and procedure manuals, trouble shooting guide, recent updates of job related information (e.g. - messages from the supervisor) diagrams, charts, circuit diagrams and tables. Using a network to access this database, information could be updated in one place and accessed by many workers. For example, even a simple aircraft can have 100,000 pages or more in its operating manual, and due to operational changes and upgrades, half of these pages might become obsolete every six months, thus the advantage of networked wearable systems allows user and the intelligent agent to access the most up to date information (Siewiorek, 1999). Again, this is an interactive system with the computer taking the role of an "intelligent tutor" information could be made available both in visual and auditory formats (Najjar et al, 1999). FAST uses a wireless network, and could include other support applications such as email, videoconferencing and interaction with other workers.

For workers utilizing wearable computers, the voice activated interface seems most useful as it allows for total hands free operation of the computer allowing better mobility especially when working in tight spaces. Again, the voice activated wearable computer will allow the user to handle tools with both hands, at the same time allowing the user to trouble shoot utilizing referencing diagrams, specifications with full duplex interaction with the computer.

5.3 Smart Houses

Houses are now being fully networked as users install coaxial cable and network cards and hubs. Although the option of a wireless network is also possible, it is somewhat more expensive. Lindquist (1999) describes the 'home of the future' as follows: "The 21st-Century home will be humming with data. A high-bandwidth pipe will deliver a full-time Net connection to every machine in your house. Even your phones will be in on the act by letting you check and send e-mail, organize your calendar, and pull information off the Web" (page 246).

Jini technology (based upon Java) allows different electronic appliances to form spontaneous networks (Lindquist, 1999). Jini allows every Jini-enabled electronic appliance to communicate, allowing networking between the washing machine, DVD player, printer, or whatever. Smart homes have a full time internet access, email, as well as full DVD signals in every room. Coordinated computer control over lighting, heat and air-conditioning, security and entertainment allows the user to program these functions in any combination. For example, one possibility is installing sensors on the floor which will automatically turn on a light when the user walks over this in the night going to the bathroom. In the morning one could browse the internet headlines while brushing one's teeth in the bathroom, peruse email, etc.

5.4 Smart Eye Glasses

MicroOptical Corporation (http://www.microopticalcorp.com/) offers a video display which attaches to ordinary eyeglasses and allows the user to manipulate the image through a wearable computer. The revolutionary part about this is that the "screen" is a tiny lens/mirror that sits in front of the eye, integrated into the eyeglass lens itself, and reflects the image into the eye (Lieberman, 1999). The user perceives an image which appears to be floating in space at an adjustable distance of a few feet.

5.5 Smart Cars

From a recent advertisement section in a New Yorker magazine advertisers touted the current "smart car" technology (1999). These innovations are already available and include a night vision system, internet and email access (using MicroSoft CE OS), voice activated controls for all dashboard functions, windshield information displays, GPS navigation, entertainment systems (CD-ROM, DVD, MP3) and satellite communications. There is currently an on line "Intelligent Transportation Systems" newsletter at http://www.ieee.org/its which, in addition to sponsoring conferences is an international clearing house for state of the field journal publications.

Automobiles are already full of integrated circuitry controlling such diverse functions as the break lights, breaks, transmission and complete engine management (Torgjornsen , 1999). For example, if the sensor in your transmission senses that a heavy load is being pressed upon the system, it will take action such as shifting into a lower gear.

Research is being conducted into cars that drive themselves. The ARGO is equipped with specialized computer devices for "autonomous driving", and in a test it successfully drove itself 2000 km's over Italian public roads (Broggi, 1998).

5.6 Military

The military has been a leader in developing new technologies and new uses for wearable computer systems. The military has a "Smart Modules Program" - (see web page at http://www.darpa.mil/MTO/SmartMod/index.html) which refers to enhancing the capabilities of users, or, in the words of the military, "warfighters". Of course the goal is increase the effectiveness of soldiers upon the battle field. The military is using wearable computer systems to allow the warfighters the ability to "better perceive their environment" by being able to "see, hear, and feel the electromagnetic spectrum" (see aforementioned web site). Military wearable computers utilize intelligent agents for assisting the fighter in the field, and utilize other functionality advances already presented (camera, compact size, networking, wireless technology, GPS, night vision).

6.0 Handicapped

Wearable computers show much promise for assistance to the disabled. Vanderheiden finds some irony in how the development of wearable computers will prove to be a win-win for the non-handicapped as well as the handicapped. Since the variety of uses for a wearable necessitates that the user be able to access the computer while engaging in a large variety of every day tasks (e.g. - walking, driving, dining, attending meeting, etc.) the interface to access the computer, by necessity, must be extremely flexible, elastic and simple (Vanderheiden & Law). As non-handicapped users utilize functions using voice activation or some other non-traditional interface modes, this can also support the handicapped user whose primary access might solely be just one of these interface access choices (e.g. - user's experiencing sensory impairment, physical impairment, cognitive impairment, etc.). All the wonderful functionality that wearable computing brings (hands free working, intelligent agents, networking diverse components both personal and professional, computer enhancement of our existing cognitive skills) will be especially welcomed for the disabled population. For example, currently head mounted cameras can provide specific assistance to the visually impaired (Starner, 1997). The video display picture can be manipulated so as to compensate for the visual irregularity of the user. Starner et al. have developed a wearable system which showed a 97% American sign language word recognition rate (using a limited vocabulary) via input from a small camera mounted on a baseball cap (1997). The wearable computer could also be an explicit guide to aid in navigation and location.

The key will be developing an interface which can meet all the divergent needs, of both the novice and the power user, of handicapped and non-handicapped. This interface will need to be sufficiently flexible to pull all divergent pieces of our information network together, (e.g. - home, office, automobile, cellular, etc.) as well as offer the handicapped user choices of I/O options such as audio or video. In order to meet these diverse user needs, Vanderheiden and Law speak of "layering" of the interface. In this manner the menu choices would be presented hierarchically with a simple display of the most frequently used options, and a series of submenus accessing more complex functions. Thus Vanderheiden titles his on line article as "anywhere, anytime and anyone", in terms of access specifications. This interface will need to work with a complex network which connects all the disparate appliances in the user's world. Currently there are microprocessors in coffeepots, shaver, toasters, etc., etc. According to Vanderheiden "…where there's a processor, there's a program. Where there's a program controlling the behavior of a product, there's the opportunity to introduce flexibility into the program and therefore into the behavior of the interface for the product" (Vanderheiden & Law). As we network all of our appliances, we have the opportunity to connect them under the control of a single interface, and this interface being accessible by all types and manner of users. Wearable computers are being studied as a remote interface (RI) to assist handicapped with hand and arm impairments (Ross et al., 1997). In this manner, researches hope to identify a single interface which can assist the handicapped user with a myriad of tasks (e.g. - accessing ATM) through the wearable computer interface.

7.0 Ergonomics A whole new set of application problems occur because these computers are
worn by the user. Taking the computer out-of-doors poses potential problems in-so-far as weather
(rain, temperature), and the actual hardware can be cumbersome and awkward to wear posing major challenges
to system efficacy (Mann, 1997, Baber et al., 1999). Baber et al. continues in speaking about how the
weight of the wearable components can alter the center of mass (COM) and stress the musculoskeletal
system leading to the wearer getting fatigued and/or experiencing decreased physical mobility
(thereby interfering with task performance). The headset can also present problems in terms of
restricting the visual field, causing the user to use more and larger head movements, as well as
impacting the user's COM (possible side effects of posture impact and muscle stress).

The problem of heat dissipation also poses challenges (Starner & Maguire, 1999).
Generally, as functionality and computing power increases, so does the heat generated by these components.
Starner and Maguire posit that in wearable computers, that the actual human body may be a very efficient
medium to dissipate heat generated from the wearable components. They conducted a study to investigate
the ability of a forearm mounted computer to dissipate heat through body contact and found very
efficient heat dissipation in wearing the forearm computer (see diagram below) (pg. 11).




The advent of "smart clothing" is taking on new dimensions with wearable
computing as designers attempt to address problem areas of weight and component distribution, making
personal computing "more personable" (Mann, 1996, Mann, 1997a). 'Smart clothing' is made up of "smart
fabric", where the use of conductive textiles allows for data lines, power distribution and sensing
circuitry to be incorporated directly into wash and wear clothing (Post & Orth, 1997). Post
and Orth examined the natural conductive value of silk thread. Running this thread through a thin
sheath of copper foil increased conductivity, was still able to withstand high temperatures,
could be washed as with any clothing and could be sewn and manipulated as with any standard cloth
material. They found that the only real concern was keeping the contacts from touching, but found
that layering or coating the material worked fine in controlling this. Metal 'clothing snaps' can
be used to attach extra components and can be used to add various functionality based upon what is
attached to the garment (thus they coin the "smartkerchief" accessory). Additionally, conductive
fabric can be sewn to act as an input keyboard, being sensitive to touch in a row-and-column-addressable
structure.

Shoes have been examined as a method of creating power in that several
watts of power are generated in an ordinary gait, and this power might be used for other peripheral
devices worn on the user's body (Post et al., 1997). Additionally, shoes (specifically sneakers in
this example) can be used both as a power source (generated when walking) and also as a communication
portal where messages can be sent out through the shoe over a "personal area network" using low
frequency 'near-field' communication into a ground transponder which is hard wired into a free
standing computer or other network (see Personal Area Network section) (Zimmerman, 1996) .



Future Research

Future research and development issues include the following:

1) Although progress is being made daily, there is continued need to manage the still limited resources of wearable computers (eg. - memory, speed, etc.), based upon the applications to be run within, the varied environments used in, and with many different types of users.

2) Ergonomic research - best method to utilize components, eg - the head set such that it is comfortable and is helping and not distracting the user from the task

3) In terms of Personal Area Networks, to increase the speed at which they can run (using the body as the networks medium); also to increase to bps speed of data transfer (within the human bus).

4) In terms of affective computing - there needs to be more light weight efficient and easily powered computer apparatus which is comfortable to wear. Also there is a need for sensors which can measure small gradients of bio-activity in a very accurate manner. Lastly there is a need to have much greater understanding of how the emotions express themselves in physiological and measurable manners.






Bibliography:

Baber, C., Knight, J., Haniff, D. & Cooper, L., "Ergonomics of Wearable Computers", Mobile Networks and Applications, 4:1, March 1999, pages 15 - 21

Bass, L., Kasabach, C., Martin, R., Siewiorek, D., Smailagic, A., Stivoric, J., "The Design of a Wearable Computer", CHI '97, March 1997, pages 139 - 146

Bass, L., Siewiorek, D., Smailagic, A., Stivoric, J., "On Site Wearable Computer System", CHI '95 Mosaic of Creativity, May 1995, pages 83 - 85

Broggi, A., "The MilleMiglia in Automatico Tour", 1998, on line information at http://millemiglia.ce.unipr.it/ARGO/english/index.html

Ditlea, S., "Cyborg for a Day", Popular Mechanics, September 1999, on line article at: http://popularmechanics.com/popmech/elect/9909EFCOBP.html

Fickas, S., Korfurm, G., Segall, Z., "Software Organization for Dynamic and Adaptable Wearable Systems", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Hughes, M., "Small OSes For Appliances: Who's moving in the Right Direction?", OS News, November 8, 1999, www.osnews.com/features/11.98/small.html
Lieberman, D., "Heads-up display can be built into eyeglasses", EE Times, 4/20/99, on line article at: http://www.eetimes.com/story/OEG19990420S0009

Lindquist, C., "21st Century Home - This Wired House", PC Computing, 12:11, November 1999, pages 246 - 266

Louderback, J., "WEARLOGIC Introduces the SmartWear Electronic Wallet", PCWeek On Line, May 7, 1999

Mann, S., "An Historical Account of the 'WearComp' and 'WearCam' inventions developed for Applications in 'Personal Imaging'", 1st International Symposium on Wearable Computers (ISWC '97), 1997a, http://computer.org/proceedings/8192/8192toc.htm

Mann, S., "Wearable Computing: A First Step Toward Personal Imaging", Cybersquare, 30:2, February 1997b

Mann. S., "Smart Clothing: The Shift to Wearable Computing", Communications of the ACM, 39:8, August 1996, pages 23 - 24

Matias, E., MacKenzie, I. S., & Buxton, W., "Half-QWERTY: Typing With One Hand Using Your Two-Handed Skills", Companion of the CHI '94 Conference on Human Factors in Computing Systems, 1994, pages 51-52

Najjar, L., Thompson, C., Ockerman, J., "Using a Wearable Computer For Continuous Learning and Support", Mobile Networks and Applications, 4:1, March 1999, pages 69 - 74

Najjar, L., Thompson, J., & Ockerman, J., "Using a Wearable Computer to Improve The Performance of Quality Assurance Inspectors in a Food Processing Plant", CHI '97 Workshop on Wearable Computing, Atlanta, GA, March 1997, http://mime1.gtri.gatech.edu/MiME/papers/CHI97_position_paper.html

New Yorker Magazine, December 6, 1999, pages 65 - 72

Picard, R., Healy, J., "Affective Wearables", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Post, E. & Orth, M., "Smart Fabric, or "Wearable Clothing", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Post, E., Reynolds, M., Gray, M., Paradiso, J., Gershenfeld, N., "Intrabody Buses for Data and Power", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Rhodes, B., "A Brief History of Wearable Computing", on line article at: http://wearables.www.media.mit.edu/projects/wearables/timeline.html, 1998?

Rhodes, B., "The Wearable Remembrance Agent: A System for Augmented Memory", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Ross, D., Sanford, J., "Wearable Computer as a Remote Interface For People With Disabilities", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Siewiorek, D., "Wearable Computing Comes of Age", Computer, 32:5, May 1999, pages 82-83, 87

Smailagic, A., Siewiorek, D., Martin, R., Stivoric, J., "Very Rapid Prototyping of Wearable Computers: A Case Study of Custom Versus Off-the-Shelf Design Methodologies", Proceedings of the 34th Annual Conference on Design Automation Conference, 1997, page 315 - 321

Starner, T., "Augmented Reality Through Wearable Computing", M.I.T. Media Laboratory Perceptual Computing Section Technical Report No. 397, 1997

Starner, T., Maguire, Y., "Heat Dissipation in Wearable Computers Aided by Thermal Coupling With The User", Mobile Networks and Applications, 4:1, March 1999, pages 3 - 13

Starner, T., Weaver, J., & Pentland, A., "A Wearable Computer Based American Sign Language Recognizer", 1st International Symposium on Wearable Computers (ISWC '97), 1997, http://computer.org/proceedings/8192/8192toc.htm

Torgjornsen, T., "Today's Cars are Smart!", Women Motorist on line Magazine, 1999, http://www.womanmotorist.com/MAINTENANCE/tomt/smart-cars.shtml

Vanderheiden , G., Law, C., "EZ Access Strategies for Cross-Disability Access to Kiosks, Telephones, and VCRs", on line keynote address from Trace R & D Center Industrial Engr Dept. Univ. of Wisconsin- Madison, http://trace.wisc.edu/docs/tide98_keynote_ez/keynote.html?wearable+computer

Vanderheiden, G., "Anywhere, Anytime (+Anyone) Access to the Next-Generation WWW", on line article from Trace R & D Center Industrial Engr Dept Univ of Wisconsin- Madison, http://trace.wisc.edu/docs/aaa/aaa.htm?wearable+computer

Wired Magazine, "Sony to Build Wearable PC", May 15, 1998, on line at: http://www.wired.com/news/technology/0,1282,12339,00.html

Zimmerman, T., "Personal Area Networks: Near-Field Intrabody Communication", IBM Systems Journal, 35:3&4, 1996, pages 609 - 617