Prototyping Wearables for Supporting Cognitive Mapping by the Blind: Lessons from Co-Creation Workshops Wallace Ugulino Hugo Fuks Pontifical Catholic University of Rio de Janeiro (PUC-Rio) Av Marques de São Vicente, 255 Gávea, Rio de Janeiro Pontifical Catholic University of Rio de Janeiro (PUC-Rio) Av Marques de São Vicente, 255 Gávea, Rio de Janeiro [email protected] [email protected] ABSTRACT This paper describes the co-creation workshops we carried out with three groups composed of blind users, mobility instructors, designers, and computer engineering students. The wearables prototyped by these groups combine verbalized warnings with haptic and audio feedback aiming at supporting blind persons in the task of identifying landmarks. The identification of landmarks is an essential skill required for the cognitive mapping and spatial representation. Since 2013 we have been worked on the investigation of wearables for supporting landmark identification and, thus, the cognitive mapping by the blind. The prototypes we describe in this paper were used in 82 trials in 2 empirical studies and are now being replicated for supporting mobility-training sessions. Keywords Blind Mobility, Spatial Representation, Landmark Identification, Wearable Computing. 1. INTRODUCTION The identification of landmarks plays a crucial role in the locomotion process of blind persons : it helps the orientation process, gives contextual information, helps the planning of routes, and helps avoiding obstacles and hazardous situations. According to the inefficiency and difference theories of spatial representation by blind persons , , this information is useful for the cognitive mapping. The problem in providing landmark information to the blind lies not only in detecting landmarks, but also in the way this information is presented to the blind. The negligence with human factors is considered to be one of the main causes blind persons do not adopt new assistive technologies . For instance, individuals that acquire blindness usually take some time to accept the whitecane during their period of mourning, but they finally accept it because the benefits outweigh the hassle associated with the cane. For any new technology the golden rule is to minimize the hassle and maximize the benefits. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected] WearSys'15, May 19, 2015, Florence, Italy. Copyright is held by the owner/author(s). Publication rights licensed to ACM. ACM 978-1-4503-3500-3/15/05…$15.00. http://dx.doi.org/10.1145/2753509.2753515 The masking phenomenon is a problem caused by technology that consists in a cognitive overload and/or the harmful interference of technology in the wearer's ability of sensing the environment – . Avoiding masking is one of the main challenges faced by designers of wearable devices for blind pedestrians when designing Electronic Travel Aids (ETA) and Electronic Orientation Aids (EOA). In order to develop wearables for supporting landmark identification (and, thus, cognitive mapping), this research used a method - documented in  - that combines an ethnographic study with co-creation workshops. During the co-creation workshops, blind subjects and mobility instructors worked together with designers and computer scientists to produce three different low fidelity (lo-fi) prototypes. We then used these lo-fi prototypes to build a high-fidelity (hi-fi) prototype. The hi-fi prototype was used later in 82 trials in two empirical studies (beyond the scope of this paper). This paper is organized as follows: Section 2 presents the literature review, with a discussion on the theories of spatial representation and a list of related works. Section 3 describes the observational study and the main lessons we learned with these observations. Section 4 describes the co-creation workshops, and the wearable prototypes. Finally, the conclusion and future works are discussed in Section 5. 2. THEORETICAL FOUNDATION The theories about spatial representation by the blind came from the Molyneux’s Problem . From the initial set of questions, three answers arose: deficiency, inefficiency, and difference theories , . The first dominant theory – the deficiency theory – says the blind could not acquire a spatial representation because it depends on the association between sight and other senses . Locke and Molyneux, 16th century philosophers, agreed with the deficiency theory. George Berkeley   is another early supporter of the deficiency theory. A particular experiment – conducted by William Cheselden  – on a boy born blind, who had surgery for cataracts at 14 years old, initially corroborated the deficiency theory, but many problems in the experiment brought even more discussion and the subject remained inconclusive. Later on, based on observations concerning the sight of newly born animals and babies, another theory took place – the inefficiency theory – in which the blind man is considered able to acquire spatial representation, despite the absence of visual information. The inefficiency theory, however, says this spatial representation is necessarily less efficient in blind than in sighted persons. Some known supporters of this theory are: Johannes Müller, Hermann von Helmholtz, and Adam Smith . Nowadays, difference theory is the most accepted one. It agrees with inefficiency theory concerning the blind's ability to acquire spatial representation, but diverges because it says the difference between blind and sighted person’s spatial representation is qualitatively different – and not necessarily less efficient. For instance, a blind person may prefer a longer path as long as it is easier to remember (avoiding cognitive overload). In this case, the “distance” does not seem like the best optimization function for the blind, but maybe the number of clues or landmarks in the path. Susanna Millar ,  and Simon Ungar – are amongst the researchers that support this theory. Our research draws from the difference theory. As a result, it takes as a premise the ability of acquiring spatial representation by the blind and aims at supporting this acquisition by means of wearable devices. The research on blind mobility may benefit from the framework defined by Michael Brambring , in which the problem of locomotion of blind pedestrians is divided in minor problems. Brambring divided the locomotion problem in two categories: problems of perception and problems of orientation. Perception problems are related both to the capacity of perceiving and avoiding obstacles (first minor problem), and the capacity of identifying landmarks in a path (second minor problem). Orientation problems are related both to the difficulty of orientation in near-space (spatial orientation, or the space within the reach of haptic exploration - aprox. 3ft) and far-space (geographical distances orientation) , . From the Perception category, Obstacle & Avoidance ETAs have been largely investigated, while Landmark Identification ETAs received little attention so far –. The most common assistive technology for obstacle detection is the white cane, and the most mature ETA for obstacle detection is the Ultra Cane – an electronic cane that uses vibration motors to indicate obstacles both at the level of the ground, hips and head . No specific work on Landmark Identification was found during our literature review, but some EOAs designed for navigation could help the blind identify objects that could serve as Landmarks, like works in –. Devices for navigation using GPS are the most common EOAs investigated in the Orientation category. Very often, these systems try to solve both the Spatial Orientation and Geographical Orientation problem, as the former implicates directly in the latter (this implication is also depicted by Brambring in his model). Among these systems, GéoTact  stands out for the way it communicates to the wearer about which directions to take: instead of saying “turn right” (for example), it informs directions by a metaphor of clocks – for instance, 12 o’clock means straight and 2 o’clock means a slight turn to the right. The authors do not say whether this approach is better or worse than the traditional way used by cars. Some efforts in the Orientation category used an approach of mapping images to auditory or tactile displays – (and also  for obstacle detection). This mapping approach has been consistently related to the masking problem, as the auditory pattern takes too much effort from the wearer to be understood. The mapping of images into auditory patterns results in overly complex patterns, requiring too many hours of training and also impairing the wearer’s pace of gait because they must interpret the sound before they can take action. This problem is well described by Borenstein and Ulrich after their experimental tests with NavBelt: The problem with this method lay in the fact that a considerable conscious effort was required to comprehend the audio cues. Because of the resulting slow response time, our test subjects could not travel faster than roughly 0.3 m/sec (1 foot/sec). And even this marginal level of performance required hundreds of hours of training time.  (p.1284) The mapping of images to tactile displays is also hard to interpret and may be even worse in the matter of communicating to the wearer because the skin’s sensibility varies and the most sensible areas are usually small (like fingertips). In addition, the loss of limb’s sensibility (peripheral neuropathy) is very common in blind persons when blindness comes together (or as a result) with diabetes . The harmful interference of ETAs and EOAs in the blind person’s ability to pick up environmental cues is called “masking”. The masking problem is one of the most frequent side effects of the technology because sight communicates a lot of information in a very efficient way through the optical nerve and no other sense seems to be capable of doing it so efficiently. The ideal solution is to make the blind capable of seeing, but that doesn’t seem achievable in short-term research. To the best of our knowledge, no device is capable of communicating by using the optical nerve, so researchers must rely in different “paths” for giving environment information to the brain. These different paths do not have the necessary “bandwidth”, which requires the designer to be careful when selecting what and how to communicate environmental cues to the wearer. Therefore, "masking" stands as a challenge for designers of wearable assistive technology for the visually impaired. Our research goal is to support the blind in the identification of Landmarks in order to help spatial representation acquisition. Because we have to communicate sensed information to the wearer, our research relies on the same risks of masking as the research with obstacle detection & avoidance or navigation. In order to avoid or minimize the occurrence of the masking phenomenon, we built three wearables prototyped by potential users and technical staff, as discussed in the subsequent sections. 3. OBSERVATIONAL STUDY We started this research by observing blind subjects in order to understand how they make spatial references, what elements in space are used as reference, and what elements they can't perceive. The observational study we made is classified as real life observation, non-structured, non-participant, and individual . This phase of the research lasted 8 months and started in July/2013 (selected quotes available at https://youtu.be/SxzPOXb4wzM). Throughout the whole research we had the support of the Benjamin Constant Institute (IBC), a Brazilian institution that has devoted 160 years to the education and support of the blind. The institution has its own ethical rules and the research procedure was evaluated and approved previously We were authorized to work with adult students of the rehabilitation department; most participants are late blind, but we also had some born blind students participating. All participants were asked to authorize us and sign a consent form in which they declare if and how they want to volunteer for the research, as Brazilian laws forbid us to pay for the participation of subjects in research. The lessons we learned during this observational phase are listed in Table 1. Table 1. Lessons from the observational phase of the research Lesson Blind individuals are inadequacy problems prone to postural Observed individuals Born blind, low vision, and late blind individuals Strategies for the cognitive mapping Born blind individuals and late blind Difficulties in geographical orientation Born blind individuals Born blind individuals and late blind and late blind Born blind and individuals We used pseudonyms for referencing all participants in this paper. late blind Difficulties in spatial orientation Difficulties in getting an overview of a scene Concerning postural problems, one mobility instructor said these are the first issues he works on with his students. The most common postural issues are: round shoulders, forward head and asymmetric feet position. According to Thales and , rounded shoulders are usually caused by a protective posture against unexpected encounters. The posture of low vision people is usually bad because they're trying to correct their field of vision. They usually have 'Forward Head' posture or an inclination to one side, and also develop round shoulders in order to avoid . [Thales, 30 y.o., physiotherapist and mobility instructor]. Text 1. Postural problems of the blind Blind persons use fixed elements as references, but they also use temporary cues as the smell of a grocery store, for instance. They use walls for helping them to walk in a straight line. The references are used especially for assuring they're in the right path or for changing directions, but they also count references sometimes. In that case, reference counting is preferred over step counting (Text 2). I don't like counting anything, either references or steps. Counting things makes me sleepy! (...) I think I have all these paths entangled in my 'head' [pointing at her head]. I know that at some point I'll reach references A, B, and C. I know that if I change my path I'll see other references and I know how to reach my destinations through them. I have it all in my mind. [Manu] Text 2. Orientation strategies based on references One of our blind participants declared she doesn't always prefer the shortest path. According to her, it's not always the best choice as it can be dangerous or have few references to help her orientate herself. The best choice, for her (Text 3), is a path with more references. How do I choose between two paths? If the longest path is 'better', then I choose it. Someone may say to me: common, choose the shortest..." NOOO! (emphasis) What if the shortest is too open or has no references? The best path, for me, is the one with more references. [Manu]. Text 3. Preference for paths with more references The most common references used in the mobility course are: texture of the floor, drafts, walls, fences, plants among others. Characterization Round shoulders, Forward head Asymmetric feet position Graphs, linked list, and doubly liked list Strategies for collecting references Main difficulties Step counting vs. external referencing Cognitive overload associated Difficulties in walking straight forward, Difficulties in body centric referencing Difficulties in external referencing Strategies used by blind persons 4. CO-CREATION WORKSHOP For the design phase, we used co-creation workshop sessions with participants ranging from engineering to design students, and from sighted people to blind participants, and also mobility instructors. There were 13 participants, as listed on Table 3. The workshop started with a brainstorming session on the main problems a blind person faces when walking without a sighted guide. During this session, instead of just listing ideas on the board, we decided to read ideas aloud from time to time, in order to make it possible for the blinds to participate. Table 3. Participants in the co-creation workshop Nickname (pseudo) Mary Louise Gunther B. Nathalie Vicky Carlson 51yo / Female 20yo / Male 24yo / Female 17yo / Female 25yo / Male Sighted / Blind? Born blind Sighted Sighted Sighted Sighted Bertha Brandon Vanity Rachel Green Johnny Duque 28yo / Female 19yo / Male 31yo / Female 27yo / Female 26yo / Male Sighted Sighted Sighted Born blind Sighted Jackie Ian Ryan 22yo / Female 55yo / Male 25yo / Male Late blind Late blind Sighted Age/Gender Formation Psychologist Eng. Student MSc Design Student Eng. Student DSc. Informatics Student Design Student Eng. Student Mobility Instructor Master in Literature MSc. Informatics Student Student Attorney Design Student Many problems were listed and the group was told to define guiding criteria for the design. The group decided the main guiding criteria as: Using embedded technology in wearables already worn by the blind, like watches, glasses and white canes; Keeping their hands free; Never blocking their ears with earplugs or something similar. After defining the guiding criteria, they were divided in 3 groups, each group with – at least – one blind user. The mobility instructor was told to be part of a group with only one blind person, in order to balance the number of stakeholders inside each group. Also, we chose to put at least one designer and one engineering student in each group. For the prototyping session, the method used was Blank Model Prototyping (BMP) , which requires the participation of potential users (we had 4 blind participants), and design and technology professionals. Blank Model Prototyping is a rapid role-playing technique that uses readily available art and craft materials to construct rough physical representations of a technological concept, according to a predetermined scenario. The method was used with the intent to collect potential user impressions and detailed ideas about the wearable to be built for the experimental part of this work. Although most of our participants had already worked with BMP in the past, we found it important to give them instructions on BMP to keep the group on the same page. Fig 2 shows participants during the role-playing session. landmark identification, we chose the radio-based solution: it is simpler and faster to develop than machine learning models for computer vision (as it is required in the other prototypes). A “Digital Beacon” identified each landmark, like a door, chair, entrance halls, hallways, etc. Our beacon is a tangible device made with microcontrollers (we used Arduino Mini Pro), 4 AA batteries, electronic components for voltage regulation, and a 433Mhz radio transmitter. Figure 2 shows a digital beacon, and two versions of the Smart Glasses made to communicate with these beacons. Each group produced their prototypes in 2 hours with their workgroups and then we moved on to the test phase. The roleplaying part of the method consists in using the prototypes in a role-playing session (Figure 1). Each group took 10 minutes to present each feature of their prototype. Figure 2. Digital Beacons and Smart Glasses The wearable consisted in glasses with embedded computing. For the first version, we used a 3D model for printing the glasses, but we changed to manufactured protection glasses (embedding our technology in it) because of ergonomic issues. We also built a glove and a belt with speakers and vibration motors in order to test different modes of feedback with our subjects, as shown in Figure 3. Figure 1. Testing prototypes ideas in a role-play session Many features were discussed and presented during the roleplaying sessions. For instance, for the feature “communicate new landmarks to the wearer”, each group chose a different solution: group 1 chose to use beeps and midi sounds (1 sound for each kind of landmark); group 2 decided to use sequences of vibrations with vibration motors mounted on a watch, and group 3 specified a text-to-speech solution. The technology chosen for the task of identifying new landmarks was the same in 2 prototypes: image processing on images obtained from mounted cameras. Group 2 chose a solution based on radios, which also incorporated cameras. As a result, the wearable would listen to the air trying to identify digital beacons. When it wasn’t possible to identify the digital beacons, the wearer must point the watch to a direction and wait for the analysis of the image. In order to develop and test each feature, we decided to use an iterative development process. The first step was to evaluate the viability of each solution. Because masking is a frequent phenomenon in ETAs and EOAs, we decided to build a simpler version of the prototypes for investigating the masking effect. For Figure 3. Smart Glasses, Glove and Belt The Glasses embedded a microcontroller Arduino Mini Pro 16Mhz 5v, a 433Mhz receiver module, Li-Po batteries 3.7v, voltage regulators (5v and 3.3v), a USB charging module, a WTV020-SD sound module (to provide verbalized warnings for the wearer, every time a new beacon is found), and a Bluetooth LE module (AdaFruit nRF8001) for the communication with the iOS App used to configure the wearables and the test modes. The glove and the belt were made with the same components of a Digital Beacon, but we added speakers and vibration motors for making it possible to test different feedback modes. 5. CONCLUSION AND FUTURE WORKS This paper presented our approach for prototyping wearables for supporting spatial representation acquisition by blind people. The work started with an observational study in which we learned the main problems related to the mobility of blind pedestrians. We were able to correlate the masking phenomenon (described in literature) with the testimonials of our blind participants. We then produced a high-fidelity prototype based on the lessons we got from the previous stages. This prototype is now being used in mobility-training sessions and we keep learning from them. In this way, this work contributes with an approach for prototyping wearables for blind users. Future works include a new version of the Smart Glasses that will feature a camera mounted between the lenses. The camera will be used for the identification of landmarks. Cameras need to be pointed at the object that is to be sensed, while radio-based solutions (like our beacons) have a perimeter in which the radio waves are captured by antennas. These differences may influence the wearer’s posture and it may influence his performance when finding landmarks, his posture, and also his social behavior. The investigation on these differences represents a second goal in the next steps of this research. 6. ACKNOWLEDGMENTS The National Council of Technological and Scientific Development (CNPq) supports this research under the project number #458.766/2013-5. Wallace Ugulino receives a grant from CNPq (#385535/2013-9). 7. REFERENCES  M. Brambring, “Mobility and orientation processes of the blind,” in Electronic spatial sensing for the blind, D. H. (Nijhoff) Warren and E. R. (Nijhoff) Strelow, Eds. Dordrecht, Netherlands: Nijhoff, 1985, pp. 493–508.  S. K. ANDREWS, “Spatial cognition through tactual maps,” in 1st Int. Symposium on Maps and Graphics for the Visually Handicapped, 1983, pp. 30–40.  J. F. Fletcher, “Spatial Representation in Blind Children: Development Compared to Sighted Children.,” J. Vis. Impair. Blind., vol. 74, no. 10, pp. 381–385, 1980.  E. M. BALL, “Electronic Travel Aids : An Assessment,” in Assistive technology for visually impaired and blind people, London: Springer, 2008, pp. 289–321.  R. Farcy, R. Leroux, A. Jucha, R. Damaschini, C. Grégoire, and A. Zogaghi, “Electronic Travel Aids and Electronic Orientation Aids for blind people: technical, rehabilitation and everyday life points of view,” in Conference & Workshop on Assistive Technologies for People with Vision & Hearing Impairments Technology for Inclusion, 2006, no. Brabyn 1982, p. 12.  J. a Brabyn, “New developments in mobility and orientation aids for the blind.,” IEEE Trans. Biomed. Eng., vol. 29, no. 4, pp. 285–9, Apr. 1982.  L. Kay, “A sonar aid to enhance spatial perception of the blind: engineering design and evaluation,” Radio Electron. Eng., vol. 44, no. 11, p. 605, 1974.  L. Kay, “An Ultrasonic Sensing Probe as a Mobility Aid for the Blind,” Ultrasonics, vol. 2, no. 2, pp. 53–59, 1964.  H. Fuks, H. Moura, D. Cardador, K. Vega, W. Ugulino, and M. Barbato, “Collaborative Museums: an Approach to Co- Design,” in Proceedings of the ACM 2012 conference on Computer Supported Cooperative Work - CSCW ’12, 2012, pp. 681–684.  J. Locke, An Essay Concerning Humane Understanding, 4th ed. Oxford: Clarendon Press, 1975.  G. Berkeley, An essay towards a new theory of vision. 1709.  G. Berkeley, A new theory of vision and other select philosophical writings. 1922.  W. Cheselden, “An Account of some Observations made by a young Gentleman, who was born blind, or lost his Sight so early, that he had no Remembrance of ever having seen, and was couch’d between 13 and 14 Years of Age,” Philos. Trans., vol. 402, pp. 447–450, 1728.  H. von Helmholtz, Handbook of physiological optics. 1925.  S. Millar, Understanding and representing space: Theory and evidence from studies with blind and sighted children. Oxford: Clarendon Press/Oxford University Press, 1994.  S. Millar, “Understanding and representing spatial information,” Br. J. Vis. Impair., vol. 13, no. 1, pp. 8–11, Mar. 1995.  S. Ungar and M. Blades, “Strategies for Organising Information While Learning a Map by Blind and Sighted People,” Cartogr. J., vol. 34, no. 2, pp. 93–110, 1997.  S. Ungar, M. Blades, and C. Spencer, “Visually impaired children’s strategies for memorising a map,” Br. J. Vis. Impair., vol. 13, no. 1, pp. 27–32, Mar. 1995.  S. Ungar, “Cognitive Mapping without Visual Experience,” in Cognitive mapping: past, present, and future, 2000, p. 221.  M. A. HERSH and M. A. JOHNSON, “Mobility : An Overview,” in Assistive technology for visually impaired and blind people, London: Springer, 2008, pp. 167–208.  M. A. HERSH and M. A. JOHNSON, “Disability and Assistive Technology Systems,” in Assistive technology for visually impaired and blind people, 2008, pp. 1–50.  B. HOYLE and D. WATERS, “Mobility AT : The Batcane ( UltraCane ),” in Assistive technology for visually impaired and blind peoples, London: Springer, 2008, pp. 289–321.  T. Ifukube, T. Sasaki, and C. Peng, “A blind mobility aid modeled after echolocation of bats.,” IEEE Trans. Biomed. Eng., vol. 38, no. 5, pp. 461–5, May 1991.  K. Ito, M. Okamoto, J. Akita, T. Ono, I. Gyobu, T. Takagi, T. Hoshi, and Y. Mishima, “CyARM: an alternative aid device for blind persons,” in CHI ’05 extended abstracts on Human factors in computing systems - CHI '05, 2005, p. 1483.  S. K. Bahadir, V. Koncar, and F. Kalaoglu, “Wearable obstacle detection system fully integrated to textile structures for visually impaired people,” Sensors Actuators A Phys., vol. 179, pp. 297–311, Jun. 2012.  S. Shoval, J. Borenstein, and Y. Koren, “Mobile robot obstacle avoidance in a computerized travel aid for the blind,” in Proceedings of the 1994 IEEE International Conference on Robotics and Automation, 1994, pp. 2023–2028.  C. Jacquet, Y. Bellik, and Y. Bourda, Electronic Locomotion Aids for the Blind: Towards More Assistive Systems, vol. 19. Berlin/Heidelberg: Springer-Verlag, 2006.  F. Dramas, B. Oriola, B. G. Katz, S. J. Thorpe, and C. Jouffrais, “Designing an assistive device for the blind based on object localization and augmented auditory reality,” in Proceedings of the 10th international ACM SIGACCESS conference on Computers and accessibility - Assets ’08, 2008, p. 263.  P. B. Meijer, “An experimental system for auditory image representations.,” IEEE Trans. Biomed. Eng., vol. 39, no. 2, pp. 112–21, Feb. 1992.  A. Hub, T. Hartter, and T. Ertl, “Interactive tracking of movable objects for the blind on the basis of environment models and perception-oriented object recognition methods,” in Proceedings of the 8th international ACM SIGACCESS conference on Computers and accessibility - Assets ’06, 2006, p. 111.  G. Sainarayanan, R. Nagarajan, and S. Yaacob, “Fuzzy image processing scheme for autonomous navigation of human blind,” Appl. Soft Comput. J., vol. 7, no. 1, pp. 257–264, Jan. 2007.  J. Zelek, R. Audette, J. Balthazaar, and C. Dunk, “A Stereovision System for the Visually Impaired,” University of Guelph, 1999., 1999.  A. Hub, J. Diepstraten, and T. Ertl, “Design and development of an indoor navigation and object identification system for the blind,” in Proceedings of the ACM SIGACCESS conference on Computers and accessibility - ASSETS ’04, 2004, p. 147.  S. Meers and K. Ward, “A Substitute Vision System for Providing 3D Perception and GPS Navigation via ElectroTactile Stimulation,” Int. Conf. Sens. Technol., no. November, pp. 551–556, 2005.  J. Borenstein and I. Ulrich, “The GuideCane-a computerized travel aid for the active guidance of blind pedestrians,” in Proceedings of International Conference on Robotics and Automation, 1997, vol. 2, no. April, pp. 1283–1288.  D. M. Nathan, “Long-term complications of diabetes mellitus,” N. Engl. J. Med., vol. 328, no. 23, pp. 1676–1685, 1993.  M. de A. MARCONI and E. M. LAKATOS, Fundamentos de Metodologia Científica. Atlas, 2010.  W. R. Wiener, R. L. Welsh, and B. B. Blasch, Eds., Foundations of Orientation and Mobility, Third edition : Volume I, History and Theory, Vol 1. AFB Press., 2010.  J. Arnowitz, M. Arent, and N. Berger, Effective Prototyping for Software Makers. Elsevier, 2010. .
© Copyright 2019