FOR INDEPENDENT LIVING
Report from a Working Conference November 21 & 22, 1994 Baltimore, Maryland Prepared by: The Center for Universal Design � 1995 Introduction
Advances in technology continue to make a tremendous impact on all of our lives. Personal computers, cellular phones, faxes, and the Internet have transformed how and where business is transacted in the 1990's. Home automation technology holds the potential for even greater change in the way we live by making our homes safer, more convenient, and more connected to the outside world.
Home automation refers to applications of technology to control functions in the home and communications with the outside world. Emerging home automation technology has great potential to help people with disabilities and older adults to live more independently in their own homes-improving their quality of life and reducing reliance upon others to meet daily living needs.
Historically, the promises of technology have often been unfulfilled. Sometimes this is because technology is developed for a problem which doesn't really exist. Other times, the technology doesn't work as well as advertised or is too difficult to understand, use, or maintain. In other instances, technology has fallen victim to inadequate funding or marketing. In any case, technology is dynamic and what was impossible in the past often becomes commonplace within a decade.
In order to evaluate emerging technologies for independent living, a conference was convened in November, 1994 in Baltimore, Maryland. Sponsored by the National Institute on Disability and Rehabilitation Research, U.S. Department of Education, the conference was a collaborative effort between the Center for Accessible Housing and Future Home®, the country's first demonstration home specifically designed to demonstrate home automation technologies for independent living. The conference brought people with disabilities together with leading technology experts in the areas of robotics, home electronic controls, telecommunications, computers, and voice recognition.
Objectives of the conference were to:
- Identify independent living needs of people with disabilities that may be amenable to technology solutions;
- Examine various emerging technologies to determine their applicability to independent living needs;
- Describe technology applications which may have universal appeal; and
- Identify obstacles and opportunities for widespread dissemination of promising technology applications.
Conference findings are intended to identify readily achievable technology applications that foster independent living, promote the widespread adoption of these applications, and guide further development of promising technology applications that are not readily feasible.
Conference participants were selected to provide a broad representation of consumers with disabilities, experts in various technology areas relevant to home automation, and the manufacturers of housing and home technology.
|Mr. Dave Andrews||National Federation of the Blind|
|Ms. Cindy Crouse-Martin||Governors Advocacy Council for Persons with Disabilities|
|Mr. Jim Doherty||U.S. Dept. of Education|
|Mr. Dick Duncan||Center for Accessible Housing, NCSU|
|Mr. Timothy Farr||Electronic Industries Foundation|
|Mr. Mickey Fields||Gallaudet University|
|Ms. Denise Figueroa||National Council on Independent Living|
|Mr. Nick Flagello||Leviton Manufacturing Co. (home electronics)|
|Mr. Andy Gianino||The Home Store (manufactured housing retailer)|
|Mr. Jeffrey Jerome||Future Home®|
|Dr. Michael Jones||Center for Accessible Housing, NCSU|
|Mr. Ron Mace||Center for Accessible Housing, NCSU|
|Ms. Shauna Malone||National Center for Medical Rehabilitation Research (NCMRR), NIH|
|Ms. Tricia Parks||Parks Associates (consultants in home automation)|
|Mr. Louis Quatrano||NCMRR, NIH|
|Dr. Uri Tasch||University of Maryland|
|Mr. David Ward||Future Home®|
Over the two-day conference, participants focused on four primary activities:
- a discussion of consumer independent living needs that might be addressed by home automation technology, along with the identification of existing technology and possible barriers.
- a tour of Future Home® to examine the state-of-the-art in home automation for independent living.
- presentations by technology experts on emerging technologies in relevant areas.
- a discussion of dissemination strategies to promote greater adoption of promising technology applications. The following sections provide a summary of the key points in each of these activities.
Identification of Independent Living Needs, Technology Solutions, and Possible Barriers
Independent Living Needs
- Participants generated a list of independent living needs that might be addressed through home automation technologies. The identified needs fall into the following general categories: exterior access; activities in the kitchen, bathroom, and laundry; child care; displays and controls on appliances; information processing and communication.
- Attractive personal lift and transfer devices (e.g., Hoyer-type lifts are cumbersome to use and institutional looking)
- Wheelchair lifts that are attractive, easily stored, and operable alone
- Trash removal/recycling assistance
- Mechanism to communicate with someone (e.g., delivery person) at the front door from a remote location in the home
- Warning system when outside surfaces are icy or wet
- Operable door and window locks
- Navigation system for visually impaired persons
- Kitchen storage that brings items into reach for wheelchair users
- Lifting assistance in the kitchen
- Kitchen refrigerator access, adjustable and easily pulled out shelves
- Washer & dryer access, perhaps a "Laundry Robot"
- Adjustable height toilet seat
- A cooktop heat indicator
- Automatic shutoff for cooktop so it is not left on by mistake
- Burners that glow at higher temperatures
- Ability to reach range controls while minimizing risk of burns
- Assistance in lifting an infant (e.g., from crib to lap; into a car seat)
- Infant restraint and bathing devices
- "Lockout" system to childproof dangerous areas (e.g., storage of cleaning supplies) that can be made readily accessible to wheelchair users when not locked
- A way to limit access to dangerous areas
Appliance Controls and Displays
- Membrane switches with suitable tactile feedback and proper markings for visually impaired people (e.g., Braille or large contrasting typography)
- Audible feedback on membrane panels
- Redundant feedback for all kitchen appliances
- Feedback available throughout the entire home (Is the furnace running? Is the garbage disposal on? Maybe it is a flashing display of some kind)
- Graphical user interfaces (for computers and for consumer appliances that use LCD display/control panels) are not usable by visually impaired people
Information Processing and Communications
- Downloading whole newspapers from on-line services (USA Today is currently available via voicemail)
- Closed captioned video teleconferencing
- Power backup systems
Possible Technical Solutions
Participants identified the following possible technology solutions to the needs expressed:
- Electronic door locks that require little strength or dexterity
- Motorized window locks and openers
- Voice activated intercom that uses micro-video camera and interfaces with house-wide TV system
- Global-positioning system with voice prompting to assist in wayfinding for people with visual impairments
- Automatic shut-off circuit for cooktop that shuts off heat when pot is removed or after designated (programmable) time period has elapsed
- Conveyor system that brings stored items to a height easily reachable by a wheelchair user
- Portable robot for heavy lifting (kitchen)
- Motorized refrigerator shelves that extend out of the open refrigerator
- Remote control for cooktop and oven
- Attractive personal transfer device (e.g., Hoyer-type lift) that can be easily stored when not in use
- Keypad-controlled lock for childproofing dangerous areas
- Mechanized crib to assist in lifting infant onto lap
- Raised bath tub for bathing infant from wheelchair
Appliance Controls and Displays
- Talking controls on appliances
- Central TV system with visual and audible feedback on household system functions (e.g., furnace on/off)
- Power backup systems to minimize impact of power failures
- Audible/voice feedback on graphical user interfaces
Information Processing and Communications
- Graphical user interface with audible voice output suitable for use by visually impaired people
- Internet access available with voice output
Barriers to Technology Applications
Conference participants were asked to identify some of the barriers they perceive to the adoption of the technology applications recommended by the group. The following potential barriers were identified:
- Manufacturers have difficulty identifying appropriate user needs and target-marketing products to meet these needs
- Product literature may not always describe the possible universal applications of a given product
- Much of assistive technology is not profitable because of the low volume of sales; manufacturers may view many of the recommended applications as assistive technology, with limited mass market appeal
- There is no distribution and installation network for most home automation products and systems; home automation installers are not knowledgeable about applications of their product to people with disabilities
- Manufacturers and home automation specialists are afraid of the negative stigma which may result from their product being associated with people with disabilities
- Consumers may resist usable technology as too "assistive" when they do not perceive that they need the assistance
- Consumers will not buy products that sacrifice aesthetics for function
Table of Consumer Requested
Technologies vs. Availability
The following table summarizes the current accessibility to consumers of the technology applications discussed. The first step in availability is technological feasibility. About two-thirds of the technologies mentioned appear to be definitely feasible while one-third require further study. The second step is to determine whether these technologies are available to consumers in a tangible form. For example, many are available as "off-the-shelf" products, some may be created as custom-built products from other components, and many are not available at all. Finally, accessibility of technology depends on knowledge about appropriate applications. The table below shows that few of these technologies are well known to consumers or those who might provide these items to consumers. As a result, lack of knowledge may be the greatest barrier to their adoption.
|Technology Application||Feasible||Available||Well Known|
|Electronic locks that require little strength or coordination||Yes||Yes||No|
|Voice-activated intercom for front door||Yes||No||No|
|Global positioning system for wayfinding||Yes||No||No|
|Icy walkway/ramp warning device||Yes||Yes||No|
|Motorized trash removal system||?||No||No|
|Easily operable motorized window openers||Yes||Yes||No|
|Adjustable height toilet seat||Yes||No||No|
|Robotic washer/dryer assistance||?||No||No|
|Multisensory indicator to show hot cooktop||Yes||Custom||No|
|Automatic shutoff when pot is removed or after time interval||Yes||Yes||No|
|Kitchen shelves that move to accessible heights||Yes||Yes||No|
|Portable robot for heavy lifting||Yes||No||No|
|Motorized extending refrigerator shelves||Yes||No||No|
|Remote cooktop controls||Yes||Custom||No|
|Attractive, self-storing transfer device||Yes||?||?|
|Keypad controlled lockout for dangerous areas/appliances||Yes||Custom||No|
|Mechanized crib to assist in lifting infant||?||No||No|
|Raised infant tub at wheelchair accessible height||Yes||?||No|
|Multisensory displays and controls for appliances||Yes||No||No|
|Central house status display which appears on TV||Yes||Custom||No|
|Electrical power backup system for critical technology||Yes||Yes||Yes|
|Voice recognition computer (home automation) interface||Yes||Yes||No|
|Remote health monitoring||Yes||Custom||No|
|Expanded online information services||Yes||Custom||Yes|
Tour of Future Home®
Session Two consisted of a group tour of Future Home®. The facility is a recently remodeled single-family home which includes an extensive collection of automation technology for people with disabilities and older adults. Most of the applications featured in Future Home® are commercially available. The technologies displayed at Future Home® include:
- Television-Based Operating System: All of the functions in the home may be controlled by special menu screens which appear in every television in the home. A person can activate any menu option with a common remote control.
- Voice Control: Many of the house functions are able to respond to voice commands from anywhere in the home.
- Automatic Wheelchair Sensing: This innovation automatically senses the presence of a wheelchair so that power doors may be opened, lights turned on, or other assistive technology activated.
- Intelligent Lighting: Virtually all lighting in the home can be controlled remotely from a bed, chair or wheelchair. "Intelligent lighting" turns lights on automatically when someone enters a room.
- Hands-Free Phones: Future Home® incorporates various styles of hands-free telephones providing the ability to independently make and receive calls.
- Special Universal Remote Control: This prototype device integrates all of the remote controls that a person might use into a single, portable unit with a built-in display screen. It incorporates a scanning feature that enables people to activate house functions by a single switch, eyeblink, or sip-and-puff switch.
- Security/Home Automation System: The security/automation system is specifically integrated with the power doors, power windows and lighting control. This system ensures a person never enters a dark area and makes the house look lived in when they are away. The system senses the presence of a wheelchair and opens the appropriate power door. When the resident is asleep or away, the system automatically closes power windows and disables the automatic wheelchair sensors.
- Audio Enhancement: The house-wide sound system boosts selected voice frequencies, which allows telephones and televisions to be more easily heard by people with hearing impairments.
- Video Camera System: This system allows the occupant to see who is at the door. Video cameras also allow caregivers to monitor from another room those who may need assistance.
- Warning and Reminder System: The computer system can be programmed to signal an alert when outside doors are opened (e.g., to curb wandering). Reminders can be set to provide a wake-up call and to prompt the occupant when it is time to take medication or complete other self-care or health maintenance activities.
- Adjustable Cabinetry: The height of tabletops and kitchen cabinets may be adjusted over a large range of heights, using a motorized system. Upper shelves can also be lowered manually to a more reachable height.
- Bright Full Spectrum Lighting: This lighting simulates sunlight to reduce the possibility of seasonal affective disorder for those who do not spend much time out of doors. Bright lighting is present throughout the home to aid people with visual impairments.
- Clothes Processor: This machine washes, then automatically dries clothes without transferring between machines. It also features easy to reach front controls.
- Electronic Water Controls: Future Home® demonstrates push button electronic water temperature/flow controls to allow people with limited mobility/strength to control water temperature and flow.
- Entertainment: A music and television entertainment system is available in any room and may be controlled from bed or wheelchair, or anywhere else in the home.
- Power Slide and Swing Doors: Intelligent doors allow hands-free access to the home for wheelchair users only.
- Voice Control Computing: Future Home® demonstrates how anyone can shop, bank, work, or access literature from their home simply by talking to their computer.
Expert Assessment of Emerging Technologies
Several technology experts gave presentations on the current state of the art in relevant fields of technology. Their remarks are summarized below.
RoboticsDr. Uri Tasch, University of Maryland
Dr. Tasch, an expert in the area of robotics, has developed a number of robotic projects for industry and for people with disabilities.
According to Dr. Tasch, fifteen to twenty years ago robotics was a promising area, and robotics has always been an area with great potential to assist people with disabilities. Few of the promises of robotics have been delivered, however. As a research area, robotics is in decline with reduced research funding available. In fact, robotics has in many cases become a subset of the discipline called "intelligent machines".
Dr. Tasch discussed the major technical barriers to the use of robotics for assistive technology:
- 1) Lack of sensing capability on the actuators does not allow tactile feedback to the operator which reduces the ability to perform fine motor activities and avoid crushing fragile objects.
- 2) Mechanical actuators are not as controllable as human muscles, tendons, etc. Mechanical systems seek to find an absolute position and may break an object so that they can achieve their programmed position. Biological systems have compliance built in due to the opposing muscle groups. A biological system can prioritize force, velocity, acceleration, and position based on the task at hand. A biological actuator (muscle) using the tactile feedback system provided by the nerves in the fingers will allow errors in absolute position so that pressure or acceleration will be correct.
- 3) Even if the barriers to precise and safe use are overcome, controlling and receiving feedback from a robot would be a technical challenge. This is especially true for people with disabilities since they may have limited control and/or sensing capability. Training is another issue, as it takes months of training by a nondisabled engineer to work the robot arm on the Space Shuttle.
- 4) The cost of robots is generally beyond the reach of most people. A small robot capable of some very limited household tasks currently would cost between $10,000 and $100,000.
Dr. Tasch was asked about some of the specific applications of robotics to user needs identified in Session One:
A robot for heavy lifting around the house must have fine controls and feedback or else it will not be very versatile. (Unfortunately, even if this were available, people with certain disabilities would not have the coordination to control the robot). A member of the panel noted that mechanical feeding robots exist now but there are still drawbacks. With such robots, the food must be precisely loaded, and one must be where the robot is. There is also a loss of human social interaction during the meal.
A storable personal transfer devices is feasible because it does not require the versatility of a general household robot. A system using robotics would, however, be very expensive.
An adjustable toilet seat that can be raised or lowered hydraulically is a very limited function and, therefore, very feasible. But this application probably would not require robotics technology.
Home AutomationTricia Parks, Parks Associates
Ms. Parks heads a well known market research firm that specializes in market research and industry consultation on home automation technology. Her remarks about the state-of-the-art in home automation are summarized below:
Some key points to remember about home automation:
- Home automation is not a particular product, it is a capability -- the capability to integrate lighting, telephone, security, entertainment, heating/cooling, etc.
- The goal of home automation is adapting the home to a person's needs. This is right on track with independent living functions for people with disabilities.
- If home automation products are integrated into the mainstream (i.e., universal design), the price will come down.
- Retrofit of home automation in existing homes is becoming increasingly feasible.
There are several standards emerging to allow home appliances to "talk" with each other. These standards specify the electrical levels and the language that the appliances will speak. These standards are X-10, Smart House, the Consumer Electronics Bus (CE Bus), and Lonworks by Echelon.
X-10 is an electrical method for allowing various X-10 modules to communicate through the AC power lines within the home. Typically, X-10 modules are used to replace light switches and receptacles, which allows lights and appliances to be remotely controlled from anywhere in the home. These modules can be controlled by handheld remote controls, other X-10 modules, computers or home automation systems. X-10 technology has been widely available for over 10 years, is inexpensive, and widely distributed. As it is older technology, it does not have collision detection or two-way verification, resulting in only moderately reliable communication. In addition, the X-10 has a very limited set of commands (on, off, bright, dim, all on, all off). While X-10 is widely distributed, its utility for enhancing independent living is not widely known by consumers or installers.
SMART HOUSE was developed by the National Association of Home Builders as a home automation standard. It uses a central controller (a computer) to talk with each appliance in the house. Based on the data from each appliance, the central controller makes decisions on what to do. The main drawbacks of a system like this is that it is more expensive and if the controller breaks down, you have no automation.
The Consumer Electronics Bus (CE Bus) standard was developed by the Electronic Industries Association, an association of consumer electronics manufacturers. This published standard specifies compatible wiring in the home and allows any manufacturer to build devices that can communicate with other devices on a CE Bus circuit. There is no central controller. Instead, each appliance is intelligent enough to get information from other appliances and make its own decisions about what actions it should take. The advantages of this system are lower cost, modularity, and tolerance of broken components.
The Lonworks system was developed by a company called Echelon. Like CE Bus this system does not use a central controller, instead opting to put the intelligence in each appliance. While the CE Bus is a specification which allows the manufacturers to build the item in any way which meets the specification, Lonworks consists of integrated circuits or "chips" that a manufacturer can plug into an appliance to make it compatible. So far, this standard has been installed mostly in commercial building products.
While SMART HOUSE, CE Bus and Lonworks can allow fully integrated home automation, it is not clear if these standards and products will co-exist as competitors or whether a dominant format will emerge. Typically the market for a new electronic technology doesn't grow quickly until a clear standard is established at the start (e.g., the CD player) or emerges through competition (e.g., the VHS vs. Betamax videotape).
Ms. Parks offered the following comments and recommendations concerning the independent living needs and possible technology applications generated in Session One:
A warning device for hearing impaired people that can indicate when an appliance is on can be accomplished now using X-10 modules. Many applications like this can be done with existing technology on a custom basis, using existing parts. The issue, however, is customization, and access to the necessary expertise to develop custom systems. There is currently no system that automatically provides these features.
Better displays, controls, and warning systems in appliances can be accomplished with existing technology but this requires greater awareness among manufacturers. The identified applications would require re-engineering, manufacturing and, of course, would involve additional development costs. Manufacturers need an incentive (like the promise of greater market share) to prompt their investment in these areas.
As far as a housewide warning system is concerned, this would require many different elements and, therefore, numerous manufacturers. All of the appliances would need to communicate and a control system would need to integrate them all. Systems like CE Bus, Lonworks, etc. can do this job if the appliances are compatible. A personal computer could also be used as a gateway to control the appliances if they were compatible with a standard like CE Bus.
Voice RecognitionJeffrey Jerome, Future Home
Mr. Jeffrey Jerome has been involved with voice recognition for the last eight years, first as an engineer evaluating voice recognition for military simulation systems and later as Director of Future Home®. His comments concerning the current state-of-the-art in voice recognition are summarized below:
Types of Systems
Voice recognition systems can vary along two dimensions: Speaker Independent/Speaker Dependent, and Discrete Speech/Continuous Speech. Speaker Independent recognition will recognize anyone's voice and respond to each command that is issued. It does not require a user to train the system. Speaker Dependent voice recognition responds accurately only to the voice of a specific person who has trained the system to recognize his or her voice. This training takes between 20 minutes and a few hours depending on the system. Speaker dependent systems are more reliable and currently are the only systems capable of reliably recognizing vocabularies large enough for dictation. Speaker independent systems are typically used with small vocabularies of commands like yes/no, numbers, etc.
Continuous speech systems recognize natural speech while discrete speech systems require the user to say a word at a time with a pause between words. Discrete speech systems are generally more accurate since the system is relieved of the task of determining when one word ends and the next begins. This is not an easy task when you consider how pronunciation changes depending upon the preceding and following words.
The most capable systems are Speaker Dependent-Discrete Speech and the least reliable are Speaker Independent-Continuous Speech systems. However, Speaker Independent-Continuous Speech Systems would have the broadest application, if they could be made more reliable.
Voice recognition has always been enticing, especially for people with disabilities. In general the technology of voice recognition has been disappointing because accuracy has been so low. Even a system with a 90% recognition rate gets one in 10 words wrong, which can prove frustrating, and very time consuming. For the last decade, reports have circulated about new systems coming out that significantly improve accuracy. In general, this news has been spread by people who casually observed these systems in the hands of a skilled user, who was having "a good day" with the system. In reality, voice recognition has been improving a little every year but is far from perfected. Continued improvement has come in three areas: voice recognition algorithms, user interfaces which let you correct your mistakes, and more powerful microcomputers.
In the last two years, the market has seen the introduction of Speaker Dependent-Discrete Speech products that are finally reliable and user-friendly enough to allow the user to work at a computer with acceptable speed and accuracy. Some examples of these systems are Dragon Dictate for Windows and the new IBM VoiceType. Although acceptably accurate for most users, all voice recognition systems make mistakes relatively frequently.
Now that voice recognition has reached a level of practical utility, there are other considerations like cost. Only a few years ago, a voice recognition system capable of recognizing 10,000 words for dictation cost between $4,000 and $8,000. Now a 30,000-word system costs under $1,000. The other portion of the cost equation is the computer which runs the voice recognition system. Five years ago a P.C. with enough power to run Dragon Dictate cost about $5,000. Today it costs about $1,700. In a few short years, the price has dropped to about one fourth of the original price. This cost reduction alone makes the technology accessible to many more people. Finally, with the popularity of multimedia audio, voice output feedback is readily available for use by people who are visually impaired.
Mr. Jerome had the following comments related to the recommendations generated in Session One:
Voice recognition is still not perfect and it often makes mistakes. It cannot tell the identity of the person speaking, and it still cannot self-correct its mistakes. It also operates much better in a quiet environment. It is essential that the system provide feedback to the user so that they can determine if the correct word has been recognized, and it should also provide a ready method for correcting mistakes. The ideal situation for such feedback is at a computer where a screen or digital voice can confirm what has been recognized.
Accordingly, a voice controlled door might present a reliability problem - exterior noise might trigger it, strangers might not know the appropriate command words, or a head cold might keep the unit from recognizing your voice, denying access to your own home. Computer based voice recognition is quite feasible. The environment is generally quiet, the user can wear a close microphone headset and the computer itself provides the required feedback.
Stand alone voice-activated home control systems fall somewhere in between. The environment can be noisy causing false "trips." The user must speak in a loud voice to be heard by the unit across the room to the system while the room acoustics further distort the input. Generally, they also only provide feedback in the room in which they are located.
Dissemination of Promising Technology Applications
Participants recommended a number of dissemination activities that should be undertaken immediately to encourage dissemination of promising technology applications. In addition to the operation of the Future Home® demonstration site, the following activities were recommended:
- Develop a "cookbook" of home automation solutions for people with disabilities that could be installed by any "handyperson." Each cookbook entry would describe the need addressed by the application, identify the devices, tools, materials, and skills required to implement the application, and provide source information about necessary materials or additional assistance.
- Compile a catalog of preferred home automation devices and manufacturers
- Make presentations to home automation specialists on applications to address the independent living needs of people with disabilities
- Conduct a series of "how to" short courses, for architects, home automation installers, etc. on home automation applications for independent living
- Establish a referral network of home automation specialists who can provide assistance to people with disabilities nationwide
- Continue outreach efforts targeting manufacturers to encourage product development that includes more flexible and usable design features like those identified by the conference participants
These recommended activities will be used to set priorities for work in the coming year as the Center for Accessible Housing and Future Home® continue to pursue technology solutions that work for all consumers.
Work has begun on development of the "cookbook" of home automation solutions recommended at the conference. Projected publication date is late fall 1996.