音声ブラウザご使用の方向け: SKIP NAVI GOTO NAVI

PERSONAL FREEDOM: A WEARABLE INTERACTIVE UNIVERSAL ACCESS DEVICE

David A. Ross, M.S.E.E., M.Ed. Atlanta VA Rehabilitation Research and Development Center Decatur, Georgia 30033

Abstract

For persons with disabilities, personal freedom is directly related to the ability to successfully access and interact with the world around them. The purpose of the Personal Freedom project is to develop a universal and interactive interface/access device for persons with disabilities. The investigators envision this interface as modular "wearable" hardware controlled by software that exhibits a sense of "shared responsibility" in meeting the particular needs of the user. Infra-Red (IR) and Radio Frequency (RF) coded commands and communications channels will be employed to access and control devices and to link the user to information sources.

Background

In the past, in a less mechanized society, differences in functional abilities were mediated through inter-personal relationships and a sense of shared responsibility for the successful functioning of each person in the community. Unfortunately, our society has become increasingly mechanized. While this mechanization may have been designed for purposes of convenience and efficiency, in many cases it was not designed for universal accessibility. And, in addi-tion, the new mechanisms may have completely replaced persons who would have taken responsibil-ity for the successful interactions of persons with disabilities. Push-button panels have replaced the elevator operator, automatic doors have replaced the "doorman,"Automatic Teller Machines (ATMs) are now replacing your personal bank teller, kiosks in malls are replacing the person in the information booth, kiosks in Post Offices are beginning to replace the postal clerk, kiosks in libraries are beginning to replace the "helpful" librarian, computers and word processors are replacing secretaries, and computer interfaces to the National Information Infrastructure (NII) or "information highway"are beginning to replace information clerks and government information offices. Getting needed tax advice, renewing your drivers license and registering to vote may soon all take place via public kiosks or computer interfaces. [1]

In November, 1994, the Technology Research Working Group stemming from the NIDRR Project Directors Meeting in January 1994, identified eight priority needs for persons with low vision. Seven of these eight priorities related to access issues: (1) accessing signs, directories and routing information for navigating large malls and transit plazas; (2) accessing video information; (3) accessing visual displays found in ATMs, home appliances, and other devices employing liquid crystal and light-emitting diode displays; (4) accessing advanced technology in the home that employ keypads, interactive displays, etc.; (5) for persons with both visual and auditory impairments accessing (being apprised of) warning sounds, fire alarms, sirens, bells, telephones, etc.; (6) accessing electronic information systems; and (7) designing universal interfaces for access by all persons with disabilities. [2]

The challenge is thus to design a universal interface that offers persons with disabilities equitable access to our increasingly mechanized environment so that they may successfully interact with this environment in the performance of their ADLs. Further, (1) the user should be able to "personalize" this interface to his/her own needs and abilities, (2) this personalized interface should be consistent across all activities and interactions, and (3) it should be present and available wherever and whenever it is needed. research questions 1. Can personal universal interface hardware be designed with the flexibility to accommodate the access needs of all persons with disabilities as described above, including: (a) access control of devices in the home and in the outdoor environment, including home appliances, thermostat, personal vehicle accessories, public elevators, automatic doors, street crossing buttons, toll gates, etc.; (b) access to ATMs, information kiosks, libraries, and post offices, etc.; (c) access to city street, mall and transport plaza orientation and navigation information; (d) interactive access and control of computers and computer programs; and (d) interactive access of NII information and resources? 2. Can the above described universal hardware be miniaturized to the extent that it may be unobtrusively and inconspicuously worn (or carried) by the user so as to be present wherever and whenever it may be needed? 3. What are the most appropriate hardware interfaces for each disability population and how may they best be integrated into a "wearable" device? 4. Can interactive interface software be designed that "shares responsibility" for successful interactions through (a) demonstrated respect for the person's needs; (b) open communication of device operation, procedures, warnings, problems, etc. that is phrased in a language appropriate to the user's level of understanding; and (c) an interactive structure that allows the user to learn the best means of performing a task through active dialogs and/or connects the user with a person who can help? 5. Are existing RF and IR communication links adequate to provide access to the ever-changing technological interfaces to information, devices, control systems, etc.; and if not, what is needed and can this be easily and modularity integrated into universal design hardware? 6. Can software be devised to offer interactive access and control of computing systems that employ Windows operating software, Unix X-Windows software, etc., and can the access software also keep pace with operating systems yet to be developed? 7. Can software be devised to offer access to World Wide Web communications interfaces that in addition to standard text include graphical page layouts with unconnected pieces of text, two-dimensional and three-dimensional graphics and pictures, and two and three-dimensional active sound files?

Methodology

The investigators are developing initial hardware to meet the needs of persons with spinal cord injury. However, modularity and interface generality will be maintained so that the same base hardware may be used by other disability groups. In this way, the immediate needs of a particular population can be addressed while developing a modular hardware base unit that can be employed to serve the needs of other populations as well. The operating software will be constructed in a similar fashion, in modules, many of which will be generalizable to other populations.

The hardware chosen for this design is comprised of modular wearable Personal Computer (PC) components. Employing this modular hardware gives the designers (and users) the flexibility to employ processors that suit current needs, and the ability to change to a more powerful processor when needed in the future. This hardware will also be configured for various information protocols for the implementation of many types of input and output hardware. Though voice interactive hardware is currently being developed, it will be possible to easily add in (or replace the speech output interface with) a video interface or braille interface. Also, by employing PC hardware, software development is simplified. Anyone with a desktop PC can develop software for this wearable PC device. This makes it possible for any programmer to develop and modify software for this device.

The communications hardware employed will also be modular and easily changed in the future. Initially, this hardware will include RF and IR circuits capable of controlling X-10 devices and IR remote-control devices. It will also include an implementation of the industry standard IR protocol (IRDA). Employing IRDA communications, the investigators will have the ability to monitor device operation and modify software quickly with from a desktop computer. IRDA software will also be designed to give the user the ability to access and control a desktop computer either via the serial port, or by emulating a wireless keyboard and mouse. A digital RF transceiver will be developed, offering computer network links and Internet connectivity.

Interaction with ATMs and kiosks will require the cooperation of ATM manufacturers and banks; however, the investigators are aware of talks in progress that would implement a secure IRDA link to ATMs [3]. Then, given the establishment of this protocol for ATMs, it should not be difficult to have this implemented in kiosks as well. Further, if kiosks are linked to Internet, then interacting with the kiosks could be accomplished without the need to travel to the kioskÑan obvious convenience for many people.

Interaction with home appliances is being driven by the home automation industries. This will eventually give a home PC access to appliances, thermostats, etc. Once the data protocol is established for this, it will be incorporated into the wearable interface.

Development of interactive "dialog' software will be accomplished through the use of fuzzy logic /neural network development system tools (MatLab¨). This system gives the investigators the ability to develop and test designs that employ fuzzy logic/neural networks and to then compile them into computer code that can be ported to the wearable device. In particular, this will be employed to develop a small-order "association memory" system to assist in word recognition and the determination of associated actions to take. The intent is to give the wearable device a means of learning appropriate responses from past history. In addition, this software is to give the user the ability to establish a training dialog with the wearable interface. And finally, these tools will be employed to make synthetic speech more "human" sounding, by learning how to put accents into words and phrases that better bring out intelligibility and meaning.

As wearable prototypes are developed, disabled persons from the community will be asked to "sit in" on evaluation meetings with experts and to critique device development. Subjects will also be employed to test prototypes at various stages in their development. To this end, they will be asked to take the prototypes home, use them for several weeks and provide a critique.

Software specific to the needs of spinal cord injury will be completed and evaluated first. Then development of software and hardware interfaces for persons with low vision will commence, followed by interface development for persons with hearing disabilities and persons who experience both hearing and visual disabilities. The investigators also plan to develop software/hardware for persons with cognitive disabilities. And, again, all these interfaces will be developed around the original modular hardware base and operating system.

Results of all subject testing will be analyzed relative to each subject grouping. Qualitative elements of tests will be descriptively presented through measures of central tendency and dispersion. Tabulation and graphic representations will be employed when appropriate. The objective data will be evaluated with multiple paired t-tests or repeated measures ANOVA to identify main effects and interactions of condition by device. Subjective data will form the basis for specific case studies that will complement the objective findings by identifying specific examples of advantages and disadvantages of the prototype device(s).

RESULTS

Work began in April, 1995, with the design and development of User-based VOice Interactive Control of the Environment (U- VOICE) for persons with spinal cord injury. Two working prototypes (wearable devices) have been developed that employ modular personal computer hardware from Adaptive Systems, Inc. The prototypes also employ voice-recognition/synthetic speech hardware from Voice Connexion, and an RF and IR/IRDA circuit board designed by the investigators. The RF circuit is capable of transmitting X-10 control codes. The IR circuit is able to learn and transmit entertainment system (TV, VCR, stereo, etc.) control codes, and effect IRDA protocol communications with a desktop computer.

Through voice-only interactions a person is currently able to train the interface device to recognize commands, set-up X-10 code sequences, and tell the device to learn IR codes from a hand-held remote control. Once learned, a simple voice command can then be used to control devices. Control of elevators is effected through the use of Infra-Link¨ IR codes. The wearable interface will correctly call an elevator and select panel buttons for any elevator employing the Infra-Link¨ hardware. It will also control Infra-Link¨ automatic doors and van lifts.

As of this writing, these prototypes are being tested by the investigators, and interactive speech algorithms are being developed. Professionals and persons with spinal cord injury will be invited to evaluate the prototypes in March, 1996.

DISCUSSION

The development of the suggested Personal Freedom access interface is a long term project that will require the cooperation of rehabilitation professionals, industry leaders, commercial leaders, and members of each disability group. It requires (1) a commitment to the idea that equitable access to information by all persons is desirable and possible, (2) a sense of mutual respect for people of diverse capabilities and backgrounds, and (3) a true concern for the personal freedom of all people. The face of technology is changing very rapidly and many competitors are developing technologies that seem to continuously outdate each other. This has made it very difficult predict the types of technology and communications that will prevail by the year 2000. However, it is clear that access to this technology will be imperative for the everyday performance of ADLs, and that the development of universal interfaces for persons with disabilities must proceed in spite of these uncertainties. Consequently, the universal interface hardware and software must be highly flexible and modular. The challenge is considerable, but the rewards to persons with disabilities are great, and if successful, the Personal Freedom access interface will be universally employed by persons across the spectrum of disabilities.

References

1. Core, Lyndell D. (1995) "Is this the post office?"RITIM- L@URIACC.URI.EDU. Dec 2 95. See also USPS Home Page: http://www.usps.gov.

2. Federal Register. (1994) Vol. 59, No. 222, Friday, November 18, 1994, pp. 59857-59860. 3. Vanderheiden, Gregg. (1995) "Infrared Data Meeting Notes" CSUN Friday, March 17, 1995.

ACKNOWLEDGEMENTS

Funding for this project was provided by the Department of Veterans Affairs, Rehabilitation Research and Development Service. The Atlanta VA Rehabilitation R& D Center is part of the Atlanta VA Medical Center Research Service, which provided laboratory space and equipment, administrative support and staff resources for this research. David A. Ross, Senior Biomedical Engineer Atlanta VA Medical Center Decatur, GA 30033