1998 Conference Proceedings

Go to previous article. 
Go to next article. 
Return to 1998 Conference Table of Contents 


Yehuda Sonnenblick
Jerusalem College of Technology
POB 16031
91160 Jerusalem
Voice: 972 - 2 - 6751210
FAX: 972 - 2 - 6422075
E-mail: Sonnenbl@brachot.jct.ac.il


Recently, blind individuals have begun to benefit from high-tech systems designed to aid them in orientation and navigation. In particular, it is desired to create a system dedicated to helping blind persons in spatially orienting themselves and, thereby, navigating more successfully inside a building. A prototype of such a system is described in the following paper. This system was conceived at and constructed by the Jerusalem College of Technology (JCT).

First, the problem is defined and various methods of addressing the goal of blind navigation are described and compared. Next, the concept of the system implemented at the Center is developed; this system employs infrared transmitters and receivers.

A discussion of the properties and functions of its various components follows. In addition, the system's special features are noted. Furthermore, the advantages of our approach are enumerated. Then, the results upon implementing the prototype at a Jerusalem school for blind, multi-handicapped children are presented. Finally, plans for the future are outlined.

The Problem and Its Solution

In many cases, a blind individual, especially one with additional disabilities, experiences difficulties in knowing his location, even in a familiar environment.

JCT has developed a prototype of an orientation and navigation system to aid blind and multi-handicapped blind persons to become more independent by telling them where they are located inside a given building. By enabling them to realize where they are at a particular instant or as they move about, the system also helps them to succeed in getting from one place to another independently. In addition, it is instrumental in protecting them from harm by automatically warning them when they approach a dangerous location, such as in front of a flight of stairs.

In general, navigation systems for the blind can be divided into two categories: outdoor and indoor. Both the purpose and the technology may differ for these two categories of blind navigation systems. It should be noted that navigation systems are useful also for visually impaired users, as well as for sighted users who find themselves in an unfamiliar environment.

In an straightforward attempt to facilitate navigation in a building in a simple manner, compact voice-output sign units were designed and constructed at the college.

These talking units were placed next to doors. These units were each prerecorded with a short auditory message describing the location. They are then activated by pressing a switch (like a doorbell). The message can be changed by rerecording a new message. In order to avoid accidental erasure, recording may only be performed when the previous message has been unprotected through the use of a special key.

Although these speaking signs are very easy to use and simple to produce and install, they can only be operated by persons who are capable of finding them and reaching them. A major disadvantage is that they cannot be activated from a distance, for example, by a blind user located in the middle of a corridor. Meanwhile, this problem has been addressed by others through their use of remotely activated signs employing IR or RF signals (Brabyn, Crandall & Gerry, 1993; Blekhorn & Evans, 1997).

Nevertheless, in all such signage systems, there are several limitations and disadvantages. For example, the recorded message cannot be customized for individual users.

A General Description of the System

The system developed at JCT addresses these problems. It consists of static transmitter units (12 cm x 8 cm x 4.5 cm) mounted on the ceilings along the building's corridors and a portable receiving unit for each user. The various building locations are coded into digital values, which are written into the corresponding transmitter units. These codes are continuously transmitted from the ceiling using safe, infrared beams emitted by LEDs with a typical wavelength of 880 nm. The portable receiver unit senses the beam closest to it and decodes the transmitted signal into a building location value. The receiving unit also contains a speaker module, which announces the user's location to him. The system is called PERSONA, an acronym for: personal, electronically-recorded, speech-output navigation aid.

Detailed Description: Functions, Features and Advantages

The location messages are customized for each user (or group of users) and electronically pre-recorded into the speaking unit. Thus, one child may hear a room referred to as "speech therapy room", whereas another child may hear the announcement "Sarah's room" for the same location. Thus the system meets the needs of different individuals and groups of persons. This user-friendly feature is important, especially in our application, because the cognitive levels of various handicapped children are often very different. This feature also enables the announcements to be changed and updated, whenever the character of a location changes. For example, next year the same room may be called "Rachel's room". Customizing location messages also permits speaking units to be made available in several languages. This is a valuable feature when a premises has an international clientele and hosts speakers of different languages or services diverse populations, some of whom function best in a certain language. Furthermore, the amount and level of information furnished may be varied according to the type of user. Besides mere orientation facts, other more-detailed information may be supplied. For example, one type of user, a new visitor, may hear a longer, elaborate message, such as "The Steiner music therapy room. Automatic computerized music stimulation sessions. Please enter quietly and follow the instructions." Whereas another person, using the system frequently, will prefer to hear simply "music room".

Each receiver unit is controlled by a microprocessor, for example, an 87C51 chip. The receiving unit is able to distinguish between a regular location and a dangerous one. In a dangerous location, the unit automatically warns the user of the danger; in other locations, the unit speaks only upon request. The definition of dangerous locations and regular ones is made by the client and programmed into the microcomputer along with the building "map", which ties locations to corresponding codes and messages. The speaker module in the receiver unit is based on voice recording and playback chips, which store the location messages.

Two different types of receiver units have been developed: a shoulder-positioned one and a hand-held one. Each type has its advantages. In both types, the microprocessor and batteries are worn in a convenient, light-weight (335 gm) belt pack. The sensor, which must be directed towards the ceiling, and the speaker, which should be close to one's ear, are contained in another module. It is this module which is different in the two receiver unit types. In the first type, the module (60 gm) is worn in a shoulder strap. The sensor is located at the top of this shoulder strap, so that it will "see" the incoming signals. The speaker is also located in the shoulder strap, so that it will be easy for the user to hear the messages, without disturbing his surroundings. This is facilitated by a user-adjustable volume control accessible on the belt pack. The employment of a shoulder strap to hold these two components affords the user almost total hands-free operation. When information about a location is desired, a handy switch on the belt pack is pressed. In the second type, the sensor and speaker are encased in a telephone-like, hand-held module, weighing just 95 gm. Although this module must be hand-held to one's ear like a telephone, it has the advantage of intuitive use precisely because the user perceives its operation to be similar to that of a portable telephone. The user presses a button located on this module, while holding the speaker to his ear. In this application, only the user himself hears the message; his neighbors do not hear it at all. Use of the unit is both private and inconspicuous. In addition, the sensor's direction is automatically optimized towards the ceiling, with very little interfering "shadow".

An alternative approach would be to employ light-weight earphones with the sensor on top. This also enables the sensor to receive the incoming signals nicely, permits hands-free operation, and supplies privately-heard messages. However, it has the significant disadvantages of being fairly conspicuous and slightly uncomfortable.

It should be stressed that this system has an advantage over systems based on the talking sign design in that the user does not have to search for, direct his unit to, and point it at the sign (McInerney, D'Allura & Benn, 1995). Another advantage is that the messages, being located in the portable, personal unit, can be speaker dependent, which is beneficial in certain applications, as mentioned above (for example, if the users speak various languages).

Implementation and Results

The system described above was researched and developed at JCT. It was customized for children studying at Keren Or, a Jerusalem educational institution for blind, multi-handicapped children. The navigation system prototype was then implemented at this school. The transmitting units of the system were installed in a temporary configuration, permitting flexibility during testing, and attached to the ceiling at a height of about 2.3 meters. In actual use, they would be permanently attached to the building. The sensors were activated at a height of about 1 to 1.5 meters.

Total field coverage was achieved by using single and multiple IR beams with appropriate solid angles. The horizontal distance between transmitters was about 2 meters. The maximum number of available building locations in the prototype system is 256. This is divided, for example, into up to 192 regular and up to 64 dangerous locations. Dangerous locations included the entrance to the pool, the elevators, and the staircases. When approaching such a dangerous location, the hand-held unit automatically, audibly warns the user of the danger. When navigating past a regular location, the unit speaks only when a button is pressed. This function was tested on several children. The children were successful in learning to operate the unit after just a short training session. They were then able to use it to determine their location in a corridor. Both types of receivers were tested. Most children preferred the hand-held, telephone-like module. This is in agreement with preferences expressed a priori in the MoBIC user-needs survey (Fritz, et. al., 1995).

Conclusion: Future Plans

We are now considering the formation of strategic partnerships to facilitate the technological transfer of this project (Gilden, 1997). Meanwhile, plans for an advanced system that will also direct a user from one location to another are being formulated. In addition to information on where he is located now, such a syste would supply the user with directions to his destination and other information, such as distance to various points on the route or the existence of local sites of particular interest (for example, upon passing the rest rooms, telephones, elevators, exits). Such a system has applications not only for visually-impaired and blind persons but also for reading-impaired, cognitively-impaired, or elderly users, and even for sighted persons, who wish to reach destinations in an unfamiliar structure.

Examples are large public buildings, office buildings, museums and libraries, schools and universities, airports, bus terminals and subways, shopping malls, hotels,hospitals and medical centers, etc. By creating a product that is usable by everyone, cost-effectiveness can be improved (Scadden, 1997). All visitors who need guidance to a specific destination in these settings could profit from the availability of this system. Their "disability" (being unfamiliar with their surroundings) would thus be served by such a system primarily designed for the disability of blindness. Hence, technology "for the handicapped" will be of service to all.


Blenkhorn, P. and Evans, D.G., 1997. "A system for enabling blind people to identify landmarks: the sound buoy." IEEE Trans. on Rehab. Eng. 5, 3, 276-278.

Brabyn, J., Crandall, W., and Gerry, W., 1993. "Talking signs: a remote signage solution for the blind, visually impaired and reading disabled." Proc. 15th Annual Int. Conf., IEEE Eng. in Medicine and Biology Society, 1309-1310.

Fritz, S., Michel, R., Raab, A., and Strothotte, T., 1995. "User interface design for a travel aid for blind people." Workshop on Non-Visual Graphical User Interfaces for Blind People, Software Ergonomie, Darmstadt.

Gilden, D., 1997. "Moving from naive to knowledgeable on the road to technology transfer." Technology and Disability 7, 115-125.

McInerney, R., D'Allura, T., and Benn, T., 1995. "An evaluation of talking signs:

Lighthouse staff utilization." Arlene R. Gordon Research Institute, The Lighthouse Inc., New York.

Scadden, L.A., 1997. "Technology and people with visual impairments: a 1997 update."

Technology and Disability 6, 137-145.

Go to previous article. 
Go to next article. 
Return to 1998 Conference Table of Contents 
Return to Table of Proceedings 

Reprinted with author(s) permission. Author(s) retain copyright.