SESAMONET: A Navigation Aid for Blind People

SesamoNet is an innovative and cost effective guide system for blind people. Thanks to the spread of handheld hardware and software technology, such as PDAs (Personal Digital Assistant) and smartphones, wireless communication, TTS (Text To Speech) and database software for mobile devices, we were able to build a portable, easy to use orientation and navigation aid which still has great development possibilities. The cost effectiveness is due to the recovery of RFID identity tags used for tracing after cattle slaughtering: these tags are then burrowed to create the grid used for navigation.

SesamoNet can also give the user environmental information such as nearby shops’ names presence of steps or obstacles, bus stops, tube entrances or crossings. Future projects include interaction with traffic lights and remote assistance centres. The system was developed with attention to usability issues, both software (user interface) and hardware (ergonomics).

General Description and System Overview

The system is made up by four actors:

• The tag grid

• The handheld (and Bluetooth earpiece / headsets)

• The RFID reader cane

• The navigation data server

The tag grid, by which absolute orientation is achieved, is made up of tags burrowed into the ground at about 4cm. These tags are used primary for animal identification but, after slaughtering, they can be recovered, reducing the cost of the transponder and possibly of the encapsulation, due to the fact that they are already protected.

The reader in the cane reads the tag's ID by RFID communication. It then sends this information through a Bluetooth channel to the PDA device.

The ID string and navigation data are processed by the program running on the PDA and then an output is sent through another Bluetooth channel to the headsets which finally produce the audio output that is used by the visually impaired person. Navigation data can be periodically downloaded from a server, provided that an Internet connection is available, be it GPRS or Wi-Fi.

The data server holds the database containing all the information that is used for navigation and the other features: tag position, tag sequence and all kind of environmental information. All data is organized in cells and cells are divided into sub-cells containing a smaller data subset that is easier to maintain and faster to access. The server also has an interface for client data synchronization: when users approach the border between two cells, the client automatically contacts the server asking for a new set of data.

At The STOA Experience: In the frame of the STOA Exhibition, the RFID-enabled projects for disability had their representation in SESAMONET (SEcure and SAfe MObility NETwork) and GLIDEO (GLove for the IDentification and DEscription of Objects). Both systems aim to improve the life quality of blind people and were chosen to be shown in the Exhibition for their innovative use of the RFID technology.

SESAMONET is an idea patented by the European Joint Research Centre and developed in collaboration with RFID Lab of the Sapienza University of Rome. It was born as a navigation system for blind people that could guide users, in order to design a user-friendly, accessible interface. Over time, we were able to modify and adapt the technology to incorporate features that could improve their every day life experience.

Q: Could you give us some examples?

A: It may seem strange, but the feature our test users most enjoyed during usability and accessibility tests was the ability of the system to describe the surroundings of the place where the user is. Of course, it can only describe static aspects of the environment, such as the presence of bus stops, crossings or shops outdoor and office names or apartment owners indoor.

Q: Are there any advantages over other navigation systems for blind people?

A: The most notable advantage of the system is its tremendous flexibility: for example, the system could easily be adapted to become a museum guide. Another great advantage over GPS based navigation systems is the accuracy: the position uncertainty is less than 20 cm versus about 2 meters in GPS based localisation technologies. Furthermore, RFID works as well indoors as it does outdoors.

Q: And what about GLIDEO?

A: GLIDEO stands for GLove for the IDentification and DEscription of Objects. We hope it will serve blind people to identify and get a description of tagged objects without the need for direct contact thanks to RFID technology. GLIDEO could provide all kinds of information about the displayed products in shops, or be able to supply the user with details of the colours of clothes in order to dress properly.

For more technical information, here are some further documents of interest:

• GLIDEO

• SESAMONET

• SESAMONET USABILITY

If you are interested in future developments, please visit us at http://voice.jrc.it/sesamonet/home.htm

GLIDEO A RFID SYSTEM FOR IDENTIFICATION AND DESCRIPTION OF OBJECTS

BY BLIND AND VISUALLY IMPAIRED PEOPLE

Ugo Biader Ceipidor Carlo Maria Medaglia

Marta Mei Maurizio Sabene

Alexandru Serbanati CATTID - Università “La Sapienza” - Rome, Italy

ABSTRACT - For blind and visually impaired people is quite impossible to be autonomous in the contemporary world, in which we are completely surrounded by information, but only visual information…

When looking for a product in shops, or ordering at the restaurant, or when they want to listen to the music on a CD, blind persons encounter the visual barrier of the written language, that only the help of another person can solve.

In this paper the authors describe a new system based on RFID technology and designed for trying to solve this situation of impossibility of information that afflicts blind people.

The project, called GLIDEO (GLove for Identification and DEscription of Objects), was developed by Marta Mei, as final thesis at “Politecnico of Milan” University, and now is being developed by the RFID LAB of “La Sapienza” University of Rome, for its interesting solution in the field or visual disability aids.

The GLIDEO tech-glove is a sort of technologic wearable accessory that can store essential information about objects, and then let the user listen to them, as audio messages, using RFID tags and a PDA (connected to a Bluetooth headset), on which the software and the information’s database are installed.

1. INTRODUCTION 1.a. Preponderance of Visual Information in the contemporary society and effects on the life of blind people.

“We are surrounded by objects, objects that not always speak a clear and understandable language.

One of the reasons of this difficulty is linked to the fact that in these last years we all privileged the sense of sight while the other senses were partially or totally excluded from the design-culture” .1

This happens because most part of the people are not blind or visually impaired and they naturally have in their sense of sight an instrument more operative and rapid than the “active tact” of blind, to obtain information from environment.

This natural tendency, unfortunately also typical to designers, often translates, from a practical point of view, into a wrong design of the solution for the final users.

But another way of communication does exist, one that has infinite possibilities of practical application: the communication through objects and elements that could stimulate not only the sight but also the other senses we own.

Using this kind of multi-sensorial language we don’t give information in a “sight-encoded” way, as a written text, a signal or a book, but we allow the same object to give information on itself.

This system is very concrete and incisive, not only because stimulating all senses allows for

1 Citation from: A.Colonnetti, “Plurisensorialità”, in

Ottagono n.121, 1996.

the immediate comprehension of a message, but also because the message will be under-stood also by all those people that can’t use all the senses.

There are some legislations regulating the right of disabled people to access to public spaces and information… but, for blind people, to do so, is probably very difficult or expensive. Unfortunately this means that they continue to have “barriers” to surpass not only for moving in the city or public spaces and shops, but also for recognizing and finding out information, objects, products, etc… also at home. 1.b. Recognition of objects by blind people and concerning problems.

Blind people can obtain information from the unwilling contact with objects, persons or animals, through the “passive tact”.

The same happens when they use the “active tact”, exploring the environment and using their hands to understand the shape of an object, in this case they can perceive the features and the tactile information immediately, somehow as the sight does for a seeing person.

More over, through the active tact, blind people can perceive other features of the objects as temperature, texture, weight… and though the tact has certain limits in confront of sight, it has a very important function in reveal to blind persons the world around them.

The adoption of tactile information takes place in a sequential-analytic way; while exploring, the blind gains partial data, that the brain later assembles to give them the right meaning.

Effectiveness of active exploration is strictly conditioned by the personal ability of the subject. Depending on the quantity of information collected, the possibilities of mistakes increase.

Unfortunately, what visually impaired people need more is the synthetic knowledge of the total than the analytic one of each part.

Another sense is very important in the life of blind people: the heard, that has great qualities of global, longer range discernment, and on which they rely upon for identification of objects and spaces.

From the sound of a vehicle they can identify its typology, if it’s a car, a motorcycle or a camion… from the resonance of an

ambient, many useful information about its features as geometry, dimensions, furniture… both indoor and outdoor.

But what they can obtain from the sense of hearing is more copious: it represents the primary source of information about the distance; they can approximately identify the direction and orientation of an event, the distance between the person and the event, when it begins and ends; they enhance their environmental knowledge, simply acquiring specific data that characterizes facts and events.

An important feature of hearing perception is that auditory receptors can’t separate useful vibration from useless or damaging ones.

When an ambient is full of vibrations, it produces a hearing situation that could determine confusion, pain and sometimes hurt; for this reason blind people are the most damaged when are present conditions of “audio confusion”. 1.c. State of Art

There are three technologic areas that can improve the condition of disabled people: the prevention of genetic malformation, rehabilitation and assistive technology for a complete social inclusion. The last technologies can compensate specific disabilities, innate or acquired, and are widely used as instrument of compensation of the residual abilities.

In the field of visual disabilities, the aids that help blind people in recognizing objects (in Italy) are really only a few, and have limits in memory and usability.

Analyzing all the products for visually impaired people, as scales, tactile watches, personal computers or telephones, we can notice they are all provided with vocal synthesis, so they communicate information through audio messages.

There is a great quantity of software and hardware for the accessibility of mobile phones and personal computers, indeed is very important to focalize on the fact that blind people use computers, PDA, mobile phones, organizers and electronic agenda a lot in everyday life, for example for reading a book o writing a letter, a document, to look for news and information on the web, or to memorize personal messages and notes.

1.d. Questionnaire and conclusions

Trying to analyse the issue of recognizability of products by blind and visually impaired people, the best way that was found to explore their real needs, was to submit a questionnaire to a group of 44 persons: blind and visually impaired, all members of web- groups working on the themes of accessibility and everyday autonomy.

Analysing their answers it was possible to assert that the problem of identification of products and concerning information, is very serious and important for their autonomy.

The 93% of interviewed considers positive the possibility of recognizing clothes, own objects (as compact discs, boxes, etc…) and their features, instantly and without recurring to their memory, to a very precise disposition of objects at home (with the possibility of doing mistakes), or to other persons.

The resolution of this problem will reflect on their everyday life as a greater independence and autonomy, and above all, it’ll make many of them more self-confident, strengthening their self-esteem.

In the questionnaire some questions concerned the difficulties for blind people in the shops or public spaces, and among all, these are some of the problems noticed: • nothing not visual, outside or on the door of

the shops, gives information about the typology of shop and products for sale;

• inside the shops, nothing not visual, gives information about the collocation of products, of the cash-desk, of dressing rooms, toilets, etc…

• the blind person (unless helped by someone else) has to explore inch by inch each shelf and product, without having important information as prices, discounts, colours, available sizes or features of the products …

Unfortunately, all the information in shops

and public spaces are till now given in a visual way, from the display window to the shelf’s labels, and sometimes also seeing persons have difficulty in large stores in orienteering and finding information and products.

2. GENERAL DESCRIPTION AND SYSTEM OVERVIEW 2.a. General features of the project.

The main idea of GLIDEO is about the creation of a portable electronic system that could help blind people in becoming autonomous in everyday life.

One of the barriers to autonomy is the impossibility for the blind to acquire information on objects, labels, spaces, etc... and this happens because all the information are given in a visual way.

Figure 1. Models of the components of GLIDEO system: “Manale, Bracciale and Ciondolo”

GLIDEO project is a new application of RFID technology (Radio Frequency Identification technology); using RFID, it allows blind people to identify products (as clothes, accessories, and other objects) at home and when they are in the shops.

The product consists in a portable Reader of electronic tags (RFID tags) that the user can wear as a bracelet or as a pendent.

(approx. dimensions: 140 x 60 x 14 mm). The Reader is composed of: • an electronic part (RFID reader module +

Bluetooth module + rechargeable battery). The three modules, Reader, Bluetooth and

battery are contained in a single plastic-box.

• “Manale”: this is the component that contains the aerial (antenna), that permits to read and write RFID tags, by pressing a read button. It’s always connected with the electronic part, with an electric wire. It has been designed to be comfortable and it’s wearable with only one hand, without obstructing movements and tactile perception of the user.

• One of two alternate ergonomic wearable containers for the electronic part: “Bracciale” (a bracelet with the RFID antenna situated on the palm of the hand); or “Ciondolo” (a pendent, that the user wear as a necklace, containing the electronic part and the antenna, with an extensible wire to allow a comfortable reading of the tags).

Users could be provided with: • a “personal kit” of electronics tags for the

blind user (for labelling his own items), that could contain different kinds of RFID tags: washable for clothes, adhesive for compact disks or other products, etc…

• a kit of tags for the shops: adhesive, tactile tags that the shop assistants can place in different points in the shop, to give blind people information about goods, prices, discounts, etc…

In this manner, blind people can recognize

immediately their own products as those in shops, only wearing the GLIDEO’s “tech-glove” which stores essential information about objects on RFID tags. The user can then listen to them, using a PDA (connected to a Bluetooth headset), on which the software and the informations’ database are installed. The use of GLIDEO is not restricted only to blind people: also visually impaired or older people, that don’t know the Braille, can use it! 2.b. Logical Architecture

The software is still under development, but the system will run in this way: the Reader, set to operate on demand-read- mode (holding the apposite button on the “Manale”), reads the RFID tag’s ID and sends the information, via Bluetooth, to the PDA (Personal Digital Assistant); the software on the PDA manages the received information and sends it to the headset as audio message, trough a “Text To Speech” software.

Main features of the PDA: • it contains the information’s database

relative to the personal tags of the user; • it runs the software for reading and writing

information on RFID tags; • it runs a Text-To-Speech software that make

it accessible and will transform the tag’s information into audio messages;

• it contains a Bluetooth module to connect with the RFID Reader and with the headset;

• it contains Wi-Fi module to connect to the information’s database in the shops and public spaces.

Figura 2. Functioning of GLIDEO 2.c. System Functionality Simply using the GLIDEO System, the RFID tags and an accessible PDA, the blind user can: 1 > memorize on a tag the information about

a jacket, a box or a CD, etc.. Simply placing the reader’s antenna, situated on the “Manale”, near a tag, the Reader communicates via Bluetooth to the PDA the code of the tag. With the specific RFID software on the PDA, the user writes the information he wants about the object and memorize them on the tag; then he can stick or otherwise stick the tag to the objects.

2 > In a second moment, when the user is looking for specific products, he can “read” them simply placing the Manale’s antenna near the tag on the product, and immediately listen through the earphone to the previously stored information.

3 > When in the shops, the user can read the tactile tags placed by the assistants (on the front door, on the shelves, etc…) and so all

the information about the kind of the shop, prices, discounts and features of the products are accessible. In this case the PDA will connect directly via Wi-Fi with the database of information contained in the shop’s computer.

4 > With the help of an assistant, the user can also memorize on his own tags the information about the goods just bought, so that at home they’ll be ready to be read.

3. FUTURE DEVELOPMENTS

RFID technology is still under development and tested in several application fields.

There are many applications already on the market, for example for tracing products, for logistic systems, for interactive application for sport and on toys, exhibitions and shops.

Probably, when the production of tags will be cheaper than now, and when the problems related to privacy will be clarified, RFID technology will replace magnetic bands and barcodes in identifying and tracing the products.

In the near future the use of GLIDEO will be more and more simple and widespread. Indeed, if we suppose that all the items will be tagged with RFID labels, the personal kit of tags for the blind user will actually be useless!

Blind people will be able to find out at home and in shops all that they need without an external help…

Moreover, in future, the possibility of

integrating GLIDEO’s features with another application: Sesamonet, a new system, developed by RFID LAB of “Sapienza” University of Rome, based on RFID technology and designed to help visually impaired people when they move in outdoor or indoor spaces, using an RFID cane.

Figura 3. The integration between GLIDEO and RadioVirgilio/Sesamonet. 4. REFERENCES In this document there are no precise references to the following texts, but a general processing and resume of the contents. • Abacus, 1989, “Gli ipovedenti quanti sono e

quali sono i loro problemi e le loro aspettative”, U.I.C.

• Auto ID Center “ La nuova rete. Identificazione automatica di qualsiasi oggetto in qualsiasi luogo”, da www.epcglobalinc.org

• Battezzati, L., Hygounet, J.L. “RFID. Identificazione automatica a radio frequenza. Tecnologie a applicazioni”, 2003, Biblioteca Tecnica Hoepli.

• Bliss & Crane – “The tactil perception, in American foundation for the blind research bulletin n.18, 1969.

• A.Colonnetti, “Plurisensorialità”, in Ottagono n.121, 1996.

• Condor & Muldon, 1973, “A statement of the need of blind and visually impaired individuals”, The New Outlook for the Blind n.8.

• Genensky, S., “Functional classification system of visually impaired to replace the legal definition of blindness”, Rand. Corp.

• Gibson, J.J. – “The senses considered as perceptual system”, 1966, Houghton Mifflin co.

• Gobetti, F., “Ipovisione, definizione legale ed atteggiamento delle istituzioni”, in AA.VV., “La problematica dell’ipovisione. Realtà e prospettive”, Atti del seminario dei quadri associativi, U.I.C.,1987.

• Hughes,R.K., “Orientation and mobility for the partially sighted”, International Journal for the education of the blind,n.4.

• ISTAT, “Le condizioni di salute della popolazione”, 2001, - Indagine Multiscopo sulle famiglie “Condizioni di salute e ricorso ai servizi sanitari Anni1999-2000”.

• Lauria, A., “La pedonalità urbana”, Maggioli editore.

• Libro Bianco ,2003, “Commissione interministeriale sullo sviluppo e l’impiego delle tecnologie dell’informazione per le categorie deboli” - (Ministero della salute, Ministero del lavoro e delle politiche sociali, Ministero per l’innovazione e le tecnologie).

• Mei, Marta, “Ascoltare l'abito : sistema per il riconoscimento dei prodotti moda da parte di utenti ipovedenti e ciechi mediante tecnologia RFID” ; tesi di laurea, Politecnico di Milano, 2003/04.

• Mehr, Freid, “Low vision care”, 1975, Professional Press.

• Pick, H. – “Tactual and haptic perception”, 1980.

• Scott, R.A. – “The making of the blind man”, 1969, Russel Sage Foundation.

AUTHORS Corresponding author : prof. Carlo Maria Medaglia CATTID - Università “La Sapienza” Rome, Italy Telephone: 06 49910892 prof. Ugo Biader Ceipidor CATTID - Università “La Sapienza” - Rome, Italy Telephone: +39 06 49910810 Marta Mei CATTID - Università “La Sapienza” - Rome, Italy Telephone: +39 339 7381747 Maurizio Sabene CATTID - Università “La Sapienza” - Rome, Italy Telephone: +39 320 9381402 Alexandru Serbanati CATTID - Università “La Sapienza” - Rome, Italy Telephone: +39 320 8189342

A RFID SYSTEM TO HELP VISUALLY IMPAIRED PEOPLE IN MOBILITY

Abstract: Orientation and mobility for visually impaired people are very difficult, especially in environments unknown or not designed with assistive purposes. In modern Welfare States, for guaranteeing autonomous living and social inclusion of blind and partially sighted people, independent mobility is an important objective to achieve. In this paper a novel ICT system, namely SesamoNet, relying on RFID technologies to support indoor/outdoor navigation of visual disabled is introduced. SesamoNet has been developed with the intent of giving users a non-intrusive, usable and safe way to move in urban environments. The authors approach - integrating traditional assistive technologies with wireless and RFID technologies to realize an intelligent and easy to use navigation system – is described, along with the functionalities and the physical and logical architecture of the system. A brief discussion on planned application scenarios, results from a preliminary usability tests, and ongoing effort to extend the system complete the contribution.

Keywords- Blind mobility, RFID navigation system

1. INTRODUCTION

Independent mobility of visual impaired pedestrians is difficult. Generally, to explore the space, sightless people use other senses (i.e. hearing, touch and smell) to compensate their disability.

More in details, in familiar settings they learn how to orientate in the spaces by recognising known features (points of reference). Nevertheless In unknown environments, navigation often becomes harder or even dangerous. Besides, in pre-designed environments, wherever indoor or outdoor, orientation may also be achieved by perceiving regularities in the objects around (reference lines). Actually in addiction to traditional assistive tools, as the white cane and the guide dog, also other supports in blinds mobility are diffusing, as for instance Tactile Ground Surface Indicators [1].

SesamoNet is a research project that aims to design and implement a reliable system to assist visually impaired citizens’ independent mobility in urban settings. The proposed approach foresees to augment the effectiveness and the usability of the aforementioned supports by the exploitation of innovative ICT’s (Information and Communication Technologies) hardware and software components.

More in details the SesamoNet solution intends to improve blind and low vision users’ mobility experience by coupling tactile perceptions with hearing aids. To this extent, wireless technologies (RFID and Bluetooth), hand-held devices (PDA and Smartphones) and specific system and application software for mobile device (e.g. Text To Speech, database, etc) are combined together.

Recent works To increase the mobility and/or safety of

disabled pedestrians, over years many ICT-enhanced support for mobility have been proposed in literature (e.g. in [4], [5], [6]) and various industrial patents were registered (e.g. [2], [3], [8]).

Although each system has its own peculiarities, recurrent limitations for these navigation systems are the following: • Systems relying on Global Positioning

Systems (GPS) technology have limited precision in detecting the user position, also urban canyon and indoor site are no suitable environment due to technological limits in GPS signals capturing ;

• Alternative solutions based on infrared devices needs a direct line between the device and the reference position;

• Other devices are frequently complex, expensive, and require continuous power supply.

In respect with some of these weaknesses, the adoption of Radio Frequency Identification (RFID) technology may represent a viable way to overcome them. Actually, although broad applications of RFID include those for supply chain management, security, and the tracking

of object [15], in very recent years also RFID-based guide system for blind navigation have been proposed.

Advantages of RFID technology for navigation systems are: • Small and easy to use devices • Possibility of their application in both

outdoor and indoor places • High availability and low prices of tags (that

can also be recycled from animal identification tags)

• Affordability and endurance of RFID devices in respect to damages of environmental factors or vandalism

More in details in [7], Kulyukin, Gharpure, Nicholson and Pavithran present a robotic guide for visually impaired. This robotic guide is composed by a Pioneer 2DX commercial robotic platform, a laptop connected to the platform’s controller and an RFID reader. This system locates RFID passive tags that act as stimuli for local navigation behaviours to achieve global navigation objectives.

The invention of Piotrowski [8] discloses a navigation system for the general public using RFID passive tags. The RFID tags are used as location markers e.g. in a shopping mall. A receiver capable of providing navigation instructions in accordance with a predetermined code read from such a tag is also described.

Willis and Helal [9] developed a system using a RFID tag grid. Each tag contains spatial coordinates and information sent to a reading system in the blind’s shoes. The reader communicates RFID data to a PDA via Bluetooth and a dedicated software application in the PDA helps the blind in obtaining relevant environmental information.

2. SYSTEM FUNCTIONALITY

In terms of supported functionalities SesamoNet offers various aids for the autonomous mobility of visually impaired persons: • Keeps the user inside a safe path; • Provides information about turns and

obstacles on the path; • Checks the right direction (useful if user falls

or is disoriented); • Provides general and specific

environmental information on demand; • Provides on-line help and assistance (via

GSM).

On this purpose a grid of RFID tags has been designed to provide information about the path the blind should follow in order to have a safe navigation. Such a grid, and the related acoustic signals, guide him along a route and alert him when the reader exits from the safe grid area. Actually when the user cannot easily perceive relevant points of reference, the designed system alerts the user via synthesized speech output, possibly also generating sounds to identify additional and useful reference lines.

As opposed to Willis and Helal [9], the authors did not store navigation information on the tags themselves in order to increase read speed and thus reliability. •

3. SYSTEM DESCRIPTION

As depicted in Fig. 1, a disabled interacts with the system simultaneously using three different devices: a RFID Cane Reader, a PDA, and a Bluetooth headset. Together with the tag grid, they constitute the physical architecture of the system.

Figure 1: Physical architecture of the system

The tag grid, by which absolute orientation is

achieved, is made by tags burrowed into the ground up to 4cm. The tags used by SesamoNet were previously used for animal identification [10] and, after slaughtering, they are recycled for this purpose, reducing the cost of both the transponder and its encapsulation, due to the fact they are already in a plastic or ceramics envelope.

30cm

30cm

Figure 2 Layout of RFID tags in the path In the current RFID grid the 134.2 KHz channel

is used because of the minimal radio frequency noise from interposing environment (i.e. concrete, water, etc) by which the signal is affected. This allows a reading range of up to 15cm. Transponders have associated data and ID memory blocks. The uniqueness of the number stored in the read-only ID memory block is guaranteed by the European Patent Convention (EPC) for animals’ identification. Each tag is associated with a position along the path (right, left or centre) and each position is associated with a different short sound. Fig. 2 shown how the path is organized.

In Fig.2 left arrows pointing inside the path

indicate with a specific beep the left tag border; similarly the opposite arrows indicate the right border with another beep. Central line also gives the blind relevant environment information generated with a text to speech.

The current version of the cane is close to the one used by blinds: white, light and easy to use. Inside the tip of the cane an RFID antenna is integrated. The handle hosts the RFID controller, as well as the rechargeable batteries and the Bluetooth 1.2 terminal. The reader is set to either operate on demand or in continuous read mode, while holding a button on the cane handle. In continuous read mode the battery lifetime is about three hours.

The RFID cane reader reads the tag ID and sends this number, via Bluetooth, to the PDA; the PDA software associates the received ID to a mobility information and, after converting it in a suitable message, sends it to the headset. SesamoNet software runs on any Windows CE based portable device having a Bluetooth antenna to communicate with the RFID cane. The system now works with a Bluetooth 1.2

interface but the authors are looking forward to upgrade it to Bluetooth 2.0 or ZigBee standards.

Logical Architecture The software, still under development, is

based on Microsoft .Net (but we are also developing a JAVA version) and can be divided in the following blocks: • Bluetooth Cane Connection Manager:

keep the BT connection channel open between the RFID reader and the PDA for tag ID string transmission. The user is not supposed to perform any task when navigation session starts or the BT manager is resumed after any loss of connection with the cane.

• Navigation Data Interface: it retrieves navigational data from a local database, providing the Navigation Logic with extended data related to a tag ID.

• Navigation Logic: this is the core software which handles navigation and tag data in order to provide the user with mandatory (safety-related) or on demand (environmental) navigation informations. It also checks if the direction is right and not reverted. This module ”tells” the user he is probing the central tag or one on the right or the left hand side of the path1. Eventually it can send a text string containing more complex navigation or environmental information to the TTS component [11].

• the Text To Speech component: it currently using a Loquendo (www.loquendo.com) library.

1 A sweep of the cane usually probes an arc in front of the user and it takes about one second. This means that, if three tags are probed while drawing a single arc, three output should be generated in this time. Thus we decided that spoken messages are not suitable (too long) so tones are used for the main three (and more frequent) navigation signals.

Figure 3 The overall logical architecture

4. TESTS

The analysis of the interaction of visually impaired persons with the system is important in order to have a usable system that enhances the standard cane and gives an added value to a blind person. For this reason the authors started a preliminary series of tests to check the system and its functionality with a small group of experts and blind users. Several critical points of SesamoNet have been identified and the results obtained from the test execution and from the discussion that followed it provided an input to the authors that are now calibrating the system by modifying tones and synchronizing the time when they are generated.

This evaluation activity confirmed that SesamoNet could be used as a relevant augmentation tool to provide additional information with respect to a normal cane already gives (physical objects identification). More details about these preliminary tests are in [12].

We are now planning a second usability test with a large number of blind users. This test, will be based on a standard methodology for the identification of human factors in man-machine interaction (scenario, tasks, observer, questionnaires, etc.). The management software Techsmith Morae2 will allow to evaluate the usability of the system, the error rates, and a learning curve, and highlights and

2 www.techsmith.com/morae.asp

proposes solutions for other problems that may eventually arise.

5. EUROPEAN INCLUSION POLICIES

SesamoNet project aims to be a valuable support for the independent mobility and living of visually impaired persons.

A key success factors for the adoption of the proposed system is related to initiative supporting the diffusion and availability of predisposed RFID grid in public places, especially in socially relevant ambient as for instance subway stations, crosswalks, and malls.

At European level, social inclusion policies and measures have been receiving more and more attention. In several European Councils the importance of these issues has been stressed; they have been also considered relevant aspects to take into account in the Information Society.

Through a line of continuity in the European Action and Strategic Plans (eEurope 2002, eEurope 2005, i2010-European Information Society) eInclusion and eAccessibility related activities have been priority issues.

Actually the term e-Inclusion has a double meaning; as stated in [16] eInclusion, other than prevent risks of digital exclusion from the Information Society, it also means tapping new digital opportunities for the inclusion of socially disadvantaged people.

These socio-economic goals have also influenced the EU research agenda, being continuously proposed and refined in the form of themes and challenges all over the Framework Programmes. Actually, in according with an important strategic goal of the Lisbon agenda (i2010), one of the challenges of the recently launched 7th Framework Programme explicitly address ICT-mediated support for the needs people with disabilities and elderlies tend to have. Looking at the opportunities to exploit ICT for social cohesion, promoted by the challenge “ICT systems and applications for better inclusion and independent living of all citizens”, SesamoNet addresses many of the objectives settled-up. In between of Assistive Technogy and Ambient Intelligent domain, the SesamoNet system provides:

: • user-centred ‘always-on’ mobility services

based on location-aware enhanced personalised services such as context-aware personal communications and always-available information access.

• An higher mobility of people and goods across different transport modes through the

provision of accessible and reliable information services;

• Increase personal independence, prolonging active participation in society and integrated care processes for the ageing and disabled population.

• It also realizes the integration of a number of advanced technologies, e.g. low-cost GNSS receivers, software defined radio technologies, high-accuracy hybrid positioning systems combined with dynamic navigation services, as well as technologies such as RFID and smart tags in combination with advanced sensors, communication and mobility management systems;

6. CONCLUSIONS AND FUTURE DEVELOPMENTS

Further enhancement and expansion of the current system features implies the need further investigation in many directions

Finding objects With the application of RFID tag on objects,

like keys or wallet, the cane could support the user to find them in a domestic environment.

SesamoNet could also help the blind to move in a shopping mall, while the system supplies the needed information about the tagged products on the shelves: e.g. the name, price, description, and other types of information [13].

Mapping spaces The future living space will be covered by a

large number of tags according to the characteristics of intelligent buildings and of pervasive computing. SesamoNet will be able to read every kind of environment information from these tags and help the blind to make mental maps of such spaces.

Personal assistant The under development software will allow in

the near future to easily customize the interaction environment according to user preferences and profile. The blind will be able, as it now done by GPS navigators, to select the destination and the paths, to store in the system’s memory the user preferences and past selections in terms of locations and most frequently selected paths [14].

Integration with the RFID glove In future there will be the possibility of

integrating SesamoNet with another application: GLIDEO (GLove for Identification

and DEscription of Objects) created by Marta Mei, as final thesis at Politecnico of Milan University, which is being prototyped by RFID Lab of “La Sapienza” University. This system, in order to overcome the inability of blinds in recognizing objects before tactile contact, is able to provide them audio information about objects, in a way compatible with the software of RadioVirgilio. The GLIDEO tech-glove is a sort of technologic wearable accessory that allows the user to store essential information about objects, and to listen them, using RFID tags and a PDA (connected to a Bluetooth headset), on which the software and the information’s database are installed.

Figure 4 Schematic representation of the interaction

beetween SesamoNet an Glideo

System’s applications SesamoNet system is expected to be

implemented in airports, rail or metro stations, shopping malls, universities, post offices, etc.

In order to understand and plan in the best way the industrialization phase and the commercial exploitation of the prototype, when the system will be used on a large scale, it’s necessary to investigate the economic opportunities and threats. An organization and business modelling, a market analysis, and an evaluation of possible application scenarios are necessary to know: the stakeholders, the organizational impact and possible funding resources to cover the costs of installation, management, and maintenance in a public and/or in a private environment.

ACKNOWLEDGEMENTS This research has been done in the

framework of the bilateral agreement between

CATTID and LUISS-CeRSI and partially supported by SILAB-IPSC of the EC Joint Research Centre (JRC), Oracle, Intel and Loquendo. The Authors wish also to thank all the people involved in development and testing of SesamoNet.

REFERENCES [1] Y. Kobayashi, T. Takashima, T. Hayashi, H.

Fujimoto, “Gait analysis of people walking on tactile ground surface indicators”, in IEEE Transaction on Rehabilitation Engineering, 2005 Mar; 13(1):53-59 2005

[2] L. W. Alonzi, D. C. Smith, G. J. Burlak, M. Mirowski, (1992). “Radio frequency message apparatus for aiding ambulatory travel of visually impaired persons”, U.S. Patent 5,144,294 issued Sept. 1, 1992.

[3] M.B. Hancock, “Electronic autorouting navigation system for visually impaired persons”. U.S. Patent 5,806,017 issued September 8, 1998.

[4] A. Helal, S. E. Moore, B. Ramachandran, “Drishti: An Integrated Navigation System for Visually Impaired and Disabled ”, in Proceedings of the Fifth International Symposium on Wearable Computers, 2001, 149 – 156.

[5] A. Hub, J. Diepstraten, T.Ertl, “Design and development of an indoor navigation and object identification system for the blind”, in Proceedings of the 6th international ACM SIGACCESS Conference on Accessibility and Computers , 2004

[6] H. Mori, S. Totani, “Robotic Travel Aid for the Blind: HARUNOBU-6”. In Proceedings of the Second European Conference On Disability, Virtual Reality, and Assistive Technology, Sövde, Sweden, 1998

[7] V. Kulyukin, C. Gharpure, J. Nicholson, S. Pavithran, “RFID in Robot-Assisted Indoor Navigation for the Visually Impaired”, Proceedings of 2004 lEEE/RSJ International Conference on Intelligent Robots and Systems, September 28 -October 2,2004, Sendai, Japan

[8] T. E. Piotrowski, “RFID navigation system”. EP patent 1 313 079, 2003

[9] S.Willis, S. Helal, “A Passive RFID Information Grid for Location and Proximity Sensing for the Blind User”, University of Florida Technical Report number TR04-009. http://www.cise.ufl.edu/tech_reports/tr04/tr04-009.pdf

[10] L. G. Caja; J. F. Vilaseca Vintro; C. Korn, “Ruminal bolus for electronic identification of a ruminant”, 2002

[11] M. Tatham, K.Morton, Developments in Speech Synthesis, Digital, October 2005.

[12] E. D’Atri, C.M. Medaglia, A. Serbanati, E. Panizzi, A. D’Atri, “A system to aid blind people in the mobility: A usability test and its results”, to be Telephone: +39 0332 785754published in proceedings of Second International Workshop on Mobile

Communications and Learning, Martinica, 22-29 April 2007

[13] C. Loebbecke, Modernizing Retailing Worldwide at the Point of Sale, Management Information Systems Quarterly Executive (MISQE) 3(4):177-187, 2004

[14] G. Costa, G. Manco, R. Ortale, D. Saccà, A. D’Atri, S. Za. “Logistics Management in a Mobile Environment: A Decision Support System Based on Trajectory Mining”, to appear in proceedings of Second International Workshop on Mobile Communications and Learning, Martinica, 22-29 April 2007

[15] R. Weinsten. “RFID:_A technical Overview and Its Application to the Enterprise”, in IT Professional 7(3): 27-33, May 2005

[16] Europe’s Information Society Thematic Portal. eInclusion - ICT & Inclusion. http://europa.eu.int/ information_society/policy/accessibility/eincl/index_en.htm

AUTHORS Corresponding author : prof. Carlo Maria Medaglia CATTID - Università “La Sapienza” Rome, Italy Telephone: +39 06 4991 0892 - CATTID - Università “La Sapienza” - Rome, Italy: prof. Ugo Biader Ceipidor Telephone: +39 06 4991 0810 Edoardo D’Atri Telephone: +39 349 7758055 Fabiano Ferrazza Telephone: +39 320 0536725 Alexandru Serbanati Telephone: +39 320 8189342 - European Commission – Joint Research Centre of Ispra – Institute for the Protection and Security of the Citizen: Graziano Azzalin Telephone: +39 0332 789008 Francesco Rizzo Telephone: +39 0332 789223 Marco Sironi

A system to aid blind people in the mobility: A usability test and its results

Abstract—Blind people need to become as independent as possible in their daily life in order to guarantee a fully social inclusion. Mobility means the possibility of freely moving, without support of any accompanying person, at home, in public and private buildings, and in open spaces, as the streets of the town. Mobile and wireless technologies, and in particular the ones used to locate persons or objects, can be used to realize navigation systems in an intelligent environment. Such systems open new opportunities to improve the speed, easiness, and safety of the visually impaired persons mobility. Using these technologies together with Text To Speech systems and a mobile-based database the authors developed a cost effective, easy-to-use orientation and navigation system: RadioVirgilio/SesamoNet1. The cost effectiveness is due to the recovery of RFID identity tags from cattle slaughtering: these tags are then borrowed to create a grid used for navigation. In this paper the results of an usability analysis of this guide system are presented. A preliminary experiment involving a small group of experts and a blind person is described. In order to evaluate the usability, three cognitive walkthrough sessions have been done to discuss the system’s basic functionality and to highlight the most critical aspects to be modified.

Keywords- Blind mobility; Mobile communication and devices; RFID; Usability test;

I. INTRODUCTION

A. Perceiving spaces by blind people Blind people compensate the lack of sight augmenting the

capabilities of the other senses. In this way they are able to create mental maps from sensory images [5]. In fact all the remaining senses, mainly tact and hearing, but also in some cases the olfaction, are used to replace, even if an incomplete way, the ability to perceive the spaces. In order to know an environment a visually impaired tries to build mentally the spaces by identifying obstacles around him and progressively searching environmental information.

RadioVirgilio/SesamoNet provides a suitable audio output indicating path borders, possible danger or other important information; this audio helps the visually impaired in creating his/her mental maps.

B. Mobility and orientation by blind people Orientation is the subjective ability to know the own

positioning in a space both in absolute sense and respect to the point of departure and the one of arrival; this is possible through a complex cognitive and perceptive process and by the acquisition of the sensory information coming from the environment and the body. A blind person can orient himself

1 Radiovirgilio/Sesamonet (SEcure and SAfe MObility NETwork) is a joint project between SILAB-IPSC of the EC Joint Research Centre (JRC) and the RFID LAB of the University of Rome “Sapienza”

in already known places by recognising known features (absolute orientation) or by perceiving unknown but regular objects, like buildings’ walls in outdoor situations or the angles of a room walls in indoor spaces (relative orientation). This first kind of orientation is, for a blind person, the most difficult one due to the complexity of collecting information from the surrounding environment only through tact and hearing. In order to explore unknown places a blind person usually needs to be guided by another person or a dog.

In order to move without a guide in an unknown space a blind needs to identify some points of reference that allow him to guess the optimal movements and the distance to cover to move from one place to an other one. A visually impaired decide her orientation from two types of information: the points of reference and the lines of reference [12]. Points of reference are every kind of environment information (tactile, acoustic or olfactory) a blind person easily perceive and/or remember in her memory: a visually impaired, has to knows his relative position with respects to these points and their meaning. Reference lines are based on a continuous environmental knowledge that allows a blind to orient himself and to maintain the right direction without the need of additional points of reference, example are the sidewalks, the walls, the changes of ground surface. RadioVirgilio/SesamoNet, through the voice synthesized instructions, supplies to the most important points of reference when the user cannot easily perceive these natural marks and generates sounds to identify additional and useful reference lines.

C. State of Art Several solutions have been proposed in the recent years to

increase the mobility and/or safety of disabled pedestrians.

A first system is proposed by Reppucci [1]. This system is composed by: an information broadcast service provided by a transmitting station covering the area of interest and a portable receiving station allowing the user to poll the transmitting station that deliver intelligible information about this area. In this way, an unsighted person receives relevant local information, such as the status of a traffic light signal.

Another widespread approach for outdoor navigation relies on the Global Positioning System (GPS), which is based on geostationary satellite signals. In the recent years several systems have been developed on this technology and oriented to visually impaired persons. These tools often integrated other technological services such as Geographical Information Systems (GIS). Remarkable blind navigation systems based on GPS, GIS and wireless technologies are presented in [7], [9] and [18]. Unfortunately, the resolution of a GPS localization for civil purposes is limited (few meters) and there are relevant delays in the answer when the receivers are used. Due to such

a lack of accuracy and the low real time responsiveness, the needed level of safety for a blind pedestrian cannot be currently provided by such a kind of systems.

Another example of GPS based system has been developed by Mori and Totani [13]. The system is composed by an automatic wheelchair, a GPS receiver and a portable GIS. Authors affirm that the obtained quality of mobility is better than the one provided by a guide dog, but on the other side this solution seems not easy to be extended to walking mobility.

The solution proposed in [1] is based on a portable device having a radio transceiver and fixed radio transceivers located as reference points. When a command is issued by the user, the portable device receives from the base transmitter a location message indicating the information to be synthesized. To improve the blind user orientation the portable transceiver should also include an electronic compass.

Another approach described by Hancock [6] proposes a portable navigation unit and a plurality of location beacons based on infrared emitting diodes. A direct line of sight is required in this case between the navigation unit and the beacons. A similar solution, proposed by Shojima et al. [17], uses a portable guidance apparatus and a plurality of markers which output a beam carrying location information and direction information of this beam.

Some authors [11,14,20] propose the usage of RFID passive tags for navigation system for blind people. RFID tags provide location markers and a mobile receiver provides user navigation information based on the code stored in each tag. These systems does not require a power supply for location markers and overcomes the GPS navigation limits in indoor environments. On the other side, the tags’ placement cost could be high and usually tags are not easy to adapt for an outdoor, harsh environment.

II. GENERAL DESCRIPTION AND SYSTEM OVERVIEW

The system is composed by the following parts:

1. RFID tag grid

2. RFID Cane Reader

3. PDA

Figure 1. The physical architecture composed by: a tag grid, the RFID reader

cane, and the PDA

.

Figure 2. Tags Rfid used in RadioVirgilio/SesamoNet

Passive RFID Tags -- The final tag grid, by which absolute orientation is achieved, will be made of tags burrowed into the ground up to 4cm. These tags were used for animal identification and, after slaughtering, they are recovered for this purpose, reducing the cost of both the transponder and its encapsulation, due to the fact they are already in a plastic or ceramics envelop. Figure 2 shows two typical transponders. In the current study the134.2 KHz channel was used because of the minimal noise from the environment (i.e. concrete, water, etc) by which the signal is affected. This allows a reading range of up to 15cm. Transponders have associated data and ID memory blocks. The uniqueness of the number stored in the read-only ID memory block is guaranteed by the European law for animals identification. More specifically, in this embodiment the capsule has the form of a bolus, as described in EP 0 849 989 [2] for use with ruminants such as cattle. Since RFID tags are quite cheap, the packaging is a relevant cost factor. Recycling of such animal identification tags significantly reduces the cost of the encapsulated transponders.

Rfid Cane Reader -- The RFID antenna is placed inside and near the tip of a white cane, designed for this purpose (see Figure 3), having a shape similar to the classic blind’s cane. The current version of the cane is made of plastic and needs an external RFID controller, Bluetooth interface, and batteries2.

2 For protoyping purposes we used a standalone battery and an RS232-to-Bluetottoh commercial converter which are bulky compared to the final prototype. A new model will be ready in December 2006.

Figure 3. RFID reading cane

The reader is set to either operate on demand or in continuous read mode, while holding a button on the cane handle. In continuous read mode the battery lifetime is about three hours. In the new version the cane will be closer to the one used by blinds: smaller, lighter and easier to use. It will be white, like the classic blind cane. Inside the tip of the cane, as in the prototype, an RFID antenna will be integrated. the RFID controller will be included in the handle, as the rechargeable batteries and the Bluetooth terminal that now are external devices.

PDA – System development and running usability tests were done using a HP iPAQ hw6500 PDA having a built-in Bluetooth 1.2. This device has 64MB of main memory and a Secure Digital (SD) expansion slot, used as data storage to contain the database file, in order to have the maximum main memory available to run the RadioVirgilio/SesamoNet application. Standard batteries provide at least three hours of continuous use while Bluetooth communication is on. This time will be further increased when the Bluetooth 2.0 or ZigBee interfaces will become available. RadioVirgilio/SesamoNet can also be supported by any Windows CE based portable device having a Bluetooth antenna to communicate with the Rfid cane.

A. Logical Architecture The software, still under development, is based on the

Microsoft .Net architecture (but we are also ready to start a JAVA version of the project) and can be divided in the following blocks:

• Bluetooth Cane Connection Manager: a BT connection channel is open between the RFID reader and the PDA for tag ID string transmission. The user is not supposed to perform any task when navigation session starts or

the BT manager is resumed after any loss of connection with the cane.

• Navigation Data Interface: it retrieves navigational data from a local database, providing the Navigation Logic with extended data related to a tag ID.

• Navigation Logic: this is the core software which handles navigation and tag data in order to provide the user with mandatory (safety-related) or on demand (environmental) navigation informations. It also checks if the direction is right and not reverted. This module ”tells” the user he is probing the central tag or one on the right or the left hand side of the path3. Eventually it can send a text string containing more complex navigation or environmental information to the TTS component [19].

Figure 4. The overall architecture

B. System Functionality A grid of tags has been designed to provide the needed

information about the path the blind should follow in order to be safe . Such a grid, and the related acoustic signals, guides him along a route alerting him when he exits the safe grid area. This system is a great improvement compared to the classical assisting technology for blind people’s mobility, like the white cane, the guide dog, and the tactile ground surface indicators

3 A sweep of the cane usually probes an arc in front of the user and it takes about one second. This means that, if three tags are probed while drawing a single arc, three output should be generated in this time. Thus we decided that spoken messages are not suitable (too long) so tones are used for the main three (and more frequent) navigation signals.

[10,16]. RadioVirgilio/SesamoNet offers various aids for the autonomous deambulation of visually impaired:

• Keep the user on a safe path

• Information about turns and obstacles in the path

• Checking of the right direction (useful if user falls or is disoriented)

• General and specific environment information on demand

• Check and trace his position in the tag grid

• On-line help and assistance (via GSM)

III. THE USABILITY TEST

A. Usability analysis

A deep analysis of the interaction of visually impaired persons with the system is fundamental in order to have a usable system that enhances the standard cane and gives an added value to the blind person. For this reason, we decided to study the usability of the system, analyzing the different functionalities, both with an expert-based study and with real users.

First, we performed three cognitive walkthrough sessions (Fig 5), in order to discuss the system’s basic functionalities and to highlight the most critical aspects to be modified or to be tested with users.

The first two walkthrough sessions were participated by the authors plus some students, while, in the third one, a blind person joined the group. In all the walkthrough sessions one or more participants executed some basic tasks within a typical scenario, while the other participants observed; during this execution, as well as after it, a discussion arose among participants with the goal of:

• assess tasks completion

• analyze encountered problems, and single and recurring errors

• make an expert usability evaluation of RadioVirgilio/SesamoNet basic features

• propose and discuss solutions

• discuss about further testing of the same or similar tasks (preparation for usability tests)

In the scenario in which the tasks were performed, a student (the blind person in the third session) tried to orientate and find his way using RadioVirgilio/SesamoNet along the path represented in Fig. 6. Each student was taught about the system tones and the tags disposition in the path; they were then instructed to approach the RadioVirgilio/SesamoNet pathway by scanning with the cane and moving on, keeping the cane at an appropriate distance from the floor and trying to find a central tag and placing over it.

Figure 5. Task execution during the cognitive walkthrough

Figure 6. The grass like carpets as they were disposed on the room floor.

Only the two gray carpets were equipped with RFID tags.

We positioned 5 grass-like carpets in the room. Two of them, the ones represented in grey in fig 6, were equipped with RFID tags that were configured as a path with starting point on the left, with a straight part followed by a 90° right turn. We juxtaposed the other inactive carpets to them in order to avoid tactile feedback that the student could exploit to identify the path. The visually impaired user confirmed that it was not possible to identify the path border by touching, as he could not find the carpet junctions due to the grass-like consistency of the carpets. The disposition of tags in the path is reported in Fig. 7, the direction of arrows representing the configuration of sound and messages for each tag (left border, right border, central path and direction, right turn). Three tasks were executed and discussed in each cognitive walkthrough session:

Fig 7. Layout and configuration of RFID tags in the path.

• Task 1 was related to finding the path in the room. The

student was first deliberately disoriented, then he was asked to find the path by just hearing the tones of the RadioVirgilio/SesamoNet system. His starting position (which he was not aware of) was either on the left or on the right of it, and he did not know where the path was located (either in front of him, or rear, or aside).

• Task 2 was about orientating in the path, i.e. finding path direction and identifying left and right borders.

• Finally in task 3 the student was positioned at the starting point of the RadioVirgilio/SesamoNet path and he was asked to move along towards the exit point.

B. Results Several information came both from the task execution and from the discussion that followed it. Results can be organized into different areas.

1. System tones. Participants noticed that the system tones and messages were not appropriate for several reasons: i) the message “go straight on” as central tag tone is too long and it is repeated too often due to the cane usage (see point 3 below); ii) right border tone is too low; iii) left border is not advertised in the curve as it is superseeded by the ‘turn right’ message, and it is likely to exit the path in that case; iv) the message “you are going backwards”, that is pronounced when the user reads tags which he already passed-by, should be delayed to after some occurrences of backward tags, in order to avoid false alarms. It then should be repeated only two or three times; v) the blind user suggested to use more tones, which could be learned by the user, “as a sort of ‘road sign’” (as he said).

2. Cane length. The cane length (that should be proportionate to user height) highlighted some problems related to tags finding and to the right turning: in fact, if the cane is too long (and that was the case for all the three users), it is not possible to scan very close

to the user body; one of the students in task 2, standing with her body on the right border tags, could not hear their tones and she only found the other two lines of tags; the blind user, in task 3, turned right too early because the cane was sensing tags too far away and thus the system pronounced the “turn-right” message;

3. Cane usage. Cane constituted a problem for students, as they tend to move it too fast, loosing some tags; in some cases the system played the tone related to a new tag before the tone of the previous was finished, leading to unrecognition of border positions.

4. Borders vs. central path. It was interesting to note that while the border signals were used to locate the path direction (task 2) they were not so used to follow the path in task 3. The blind user, especially, followed the central line of tags and was able to follow the path walking quickly; for this reason we repeated this task three times with him just using one tag line and also placing an obstacle on the path; lateral tags allow for redundancy and guarantee for fault tolerance, they can help to find again a lost path and they can help in closed building (that have no lodges) and in case an obstacle (e.g. a bag) is placed on the central path line; in other cases, however, it seems that central line tags could be embedded in the lodges with no need for lateral tags;

5. Look ahead. Synchronization is quite important when a turning approaches. We discussed two possible solutions to the problem of turning too early or too late based on the possibility of explicitly saying something like “two steps from now, turn right” or configuring the turn right messages on tags placed in the final part of the straight path only, in order to let the user turn as soon as he receives the message, using it as a virtual wall.

6. “Augmented reality”. The whole discussion confirmed that RadioVirgilio/SesamoNet is to be used as an augmentation of the information that the normal cane already gives (physical objects identification) and that its usage must be integrated in the cane usage. User tests will stress the analysis of this aspect.

Based on this input, we are now “calibrating the system” (modifying some system tones and the moments when they must be played) and we are planning a usability test with blind users that will be conducted in the next weeks and that is aimed to assess the validity of our assumptions and of the whole system. This test, that will be based on standard methodology (scenario, tasks, observer, questionnaires, etc.) and with the management software Techsmith Morae (www.techsmith.com/morae.asp), will allow to evaluate the usability of the system, the error rates, and a learning curve of RadioVirgilio/SesamoNet, and to highlight and propose solutions for potential other problems that may eventually arise.

IV. FUTURE DEVELOPMENTS The RFID technology is still under development and test in

several application fields. There are many applications already in the market (e.g., for tracing the products). Probably RFID technology, when production costs will be lower than now, will replace magnetic bands and barcodes in identifying and tracing the products. In the near future there could be additional applications of the RadioVirgilio/SesamoNet. For example blind could also be able to buy products in shops without the need of external aid: the cane, will help them to move in the shopping mall, while the system (according to what is suggested in [4]) will supply the necessary information about the products on the shelves: e.g. the name, price, description, and other types of information.

In a domestic environment, RFID tags, combined with the cane, could also be useful for visually impaired to find objects. Applying tags for example on keys or on the wallet they could be able to find them in a easy way [8].

A. Mapping spaces

Incrementing the number of the “tagged” paths will allow the e-cane to be a useful resource for a blind in exploring the living spaces. This system could provide a complete mapping for every kind of space and the visually impaired could be able to have mental maps of the spaces around him.

B. Personalized Assistant

A software for the management of the system is under development and will allow it’s customization according to user preferences and choices. It will be able, for instance to define the preferred path, as it is now in GPS navigators . Moreover it will be possible to store in the system’s memory the user preferences and past selections in order to automatically suggest the destinations and the most frequently used paths.

REFERENCES [1] L. W. Alonzi, D. C. Smith, G. J. Burlak, M. Mirowski, (1992). “Radio

frequency message apparatus for aiding ambulatory travel of visually impaired persons”, U.S. Patent 5,144,294 issued Sept. 1, 1992.

[2] L. G. Caja; J. F. Vilaseca Vintro; C. Korn, “Ruminal bolus for electronic identification of a ruminant”, 2002 http://www.freepatentsonline.com/6474263.html

[3] S.Cardin, D.Thalmann and F. Vexo, “Wearable System for Mobility Improvement of Visually Impaired People”, VR Workshop on Haptic and Tactile Perception of Deformable Objects, Hannover, Germany, December 2005.

[4] H. Donker, P. Klante, P. Gorny, “The design of auditory user interfaces for blind users”. Proceedings of the second Nordic conference on Human-computer interaction, October 19-23, 2002, Aarhus, Denmark

[5] M.-A. Espinosa, S. Ungar, E. Ochaíta, M. Blades, C. Spencer, “Comparing methods for introducing blind and visually impaired people to unfamiliar urban environments”. Journal of Environmental Psychology, 18, 277-287, 1998

[6] M.B. Hancock, “Electronic autorouting navigation system for visually impaired persons”. U.S. Patent 5,806,017 issued September 8, 1998.

[7] A. Helal, S. E. Moore, B. Ramachandran, “Drishti: An Integrated Navigation System for Visually Impaired and Disabled ”, Proceedings of the Fifth International Symposium on Wearable Computers, 2001, 149 – 156.

[8] A. Hub, J. Diepstraten, T.Ertl, “Design and development of an indoor navigation and object identification system for the blind”, proceedings of the 6th international ACM SIGACCESS Conference on Accessibility and Computers , 2004

[9] Z. Hunaiti, V, Garaj, W. Balachandran, F. Cecelja, “An Assessment of 3G link in a Navigation Systems for Visually Impaired Pedestrians”, Proceedings of the 15th International Conference on Electronics, Communications and Computers, 2005, 180 -186.

[10] Y. Kobayashi, T. Takashima, T. Hayashi, H. Fujimoto, “Gait analysis of people walking on tactile ground surface indicators”, in IEEE Transaction on Rehabilitation Engineering, 2005, 13:1, pp 53-59

[11] V. Kulyukin, C. Gharpure, J. Nicholson, S. Pavithran, “RFID in Robot-Assisted Indoor Navigation for the Visually Impaired”, Proceedings of 2004 lEEE/RSJ International Conference on Intelligent Robots and Systems, September 28 -October 2,2004, Sendai, Japan

[12] O. Lahav, D. Mioduser, “Multisensory virtual environment for supporting blind persons’ acquisition of spatial cognitive mapping, orientation, and mobility skills”, Proc. 4th Intl Conf. Disability, Virtual Reality & Assoc. Tech., Veszprém, Hungary, 2002

[13] H. Mori, S. Totani, “Robotic Travel Aid for the Blind: HARUNOBU-6”. In Second European Conference On Disability, Virtual Reality, and Assistive Technology, Sövde, Sweden, 1998.

[14] T. E. Piotrowski, “RFID navigation system”. EP patent 1 313 079, 2003 [15] A. L. Reppucci, “Computer controlled information broadcasting system

for aiding blind persons in town walking”, EP patent 0 338 997, 1989. [16] M. J. Scherer, “MPT-Matching persons and technology”,

http://hometown.aol.com/IMPT97/MPT.html, 1994 [17] H. Shojima, T. Yashiki (2000). “Information processing apparatus and

pedestrian navigation system using the same”, U.S. Patent 6,259,990 issued January 28, 2000

[18] T.Strothotte, S. Fritz, R. Michel, A. Raab, H. Petrie, V. Johnson, L. Reichert, A. Shalt, “Development of dialogue systems for a mobility aid for blind people: initial design and usability testing”, ACM SIGACCESS Conference on Assistive Technologies, Proceedings of the second annual ACM conference on Assistive technologies, 1996.

[19] M. Tatham, K.Morton, Developments in Speech Synthesis, Digital, October 2005.

[20] S.Willis, S. Helal, “A Passive RFID Information Grid for Location and Proximity Sensing for the Blind User”, University of Florida Technical Report number TR04-009. http://www.cise.ufl.edu/tech_reports/tr04/tr04-009.pdf