Michael Treml, Dominik Busse, Andreas Dünser, Mike Busboom and Wolfgang L. Zagler

BrailleRing: a flexible Braille display

De Gruyter | 2020

Abstract

The “BrailleRing” is a refreshable Braille display that allows blind people to read tactile characters on the inside of a rotating ring. Dots are not displayed by moving pins, but by fixed patterns on rotating cuboids. The concept allows for a flexible line-length as well as for a more robust and easy-to-clean device, because it needs far fewer actuators than piezo-electric Braille displays, and these actuators are mechanically isolated from the Braille cells.

Introduction

With the system called Braille, it is possible for blind people to gain literacy, do complicated math and read music notation, among many other things. Each character in Braille is represented by a pattern of up to eight tactile dots that blind readers can detect with a stroke of their fingers. A single dot has a diameter of only 1.5 mm [1].

Devices that display digital text in Braille are called refreshable Braille displays. Presently used technologies for these displays have some significant drawbacks such as considerable sensitivity to dirt and moisture and, therefore, high service costs.

A major reason for these problems is the number of delicate components needed. Conventional Braille displays on the market rely on piezo-electric benders as introduced by J.F. Tetzlaff in 1981 [2]. In these devices, each dot is represented by a pin that is raised or lowered by its own actuator, adding up to a whopping 320 actuators in a standard display with 48-dot-characters.

Many alternative approaches have been discussed, including shape memory alloys [3], pneumatic or hydraulic drives [4], [5] and direct skin stimulation by electric current [6]. But none of these concepts was able to leave their early prototype phases.

Our new concept started with the idea to combine two approaches that have only been examined separately before: displaying Braille characters on a rotating disk or drum [7], [8], [9] and using movable elements with rigid dot patterns instead of moving each dot individually [10], [11].

Methods

Our initial focus was to draft and compare different possible solutions. The most promising approach was then implemented by building a simplified prototype on a scale of 1:2.5 as a first technical proof-of-concept. This was done in two diploma theses [12], [13] and resulted in a patent [14]. User involvement was limited to a single Braille expert at that time because the intellectual property (IP) was not protected yet and the involvement had to take place on a non-disclosure basis.

Afterward, user involvement became a large and still ongoing part of further development. Around 50 blind people from Austria and around five people from other countries were included in the studies. They participated at several occasions:

  1. Eleven people took part in a first round of separate interviews in 2017 to collect facts about their current display usage—mostly specifications about the devices they own, the contexts they use them in and their experiences with robustness.

  2. Nineteen people took part in semi-structured interviews to make us understand the more general context of Braille usage in everyday life, including questions about work, education and mobility. These interviews were conducted by two bachelor students.

  3. Ten people took part in studies conducted by two master students in the context of their theses [15], [16]. These user studies had less focus on open interview questions and more focus on getting precise feedback on current mock-ups.

  4. Five people took part in a focus group in Linz. The topics were mostly the same as in the students’ studies.

Several people with whom we spoke were already known by team members and were met at exhibitions and conferences. Some people contacted us after reading about our project in the media.

Having several interviews done by students had the main goal of avoiding biases in our core team after having already worked on the technology for two years. We also had the bachelor students introduced to interview methodology by an external expert.

The focus group in Linz consisted of people we only met at this event. Other participants may have been included on several occasions, e.g., the 19 people included by the master students may have contained participants also interviewed by the bachelor students. The detailed number of this overlap cannot be determined because study data were anonymized for data protection reasons.

In all non-spontaneous cases of personal user involvement, we provided at least some very simple mock-ups in real size. These were usually shown at the end of the study, so participants were not influenced by them when answering more general questions. These dummies were built to observe how users might handle the device (Figure 1).

Figure 1: Handling of a simple dummy.

Figure 1:

Handling of a simple dummy.

In the first two user studies, data were collected during the interviews by filling in the blanks in forms we had prepared. The interview was audio recorded and used to add to the notes afterward. In the master students’ studies, audio and video were recorded and the audio data were transcribed. In the final focus group, our core team collected audio data and photographs of mock-up handling and compared the gathered information with the results of previous studies.

Our further technical development was split into four work packages:

  • – Actuators,

  • – General mechanical parts,

  • – Sensors and

  • – Software

Each of these parts was improved in an iterative process and at the time of this paper writing, first steps are being taken to combine them into to a first working prototype in real size.

For the mechanical components, many parts have been built with 3D printing, using FDM as well as SLA. Electronics and the software are currently based on Arduino.

Results

The new refreshable Braille display, called the BrailleRing, displays the Braille characters on the inside of a rotating ring. While the reading finger rests at the bottom of the device, the ring rotates, letting the Braille characters glide beneath the reading finger.

The Braille characters are set by turning small cuboids with rigid dots on their lateral surfaces. On each surface, there are either two dots, a dot on the left, a dot on the right or no dots at all, representing each possible combination for a line of dots within a single Braille character. The turning of these cuboids is achieved in the upper half of the rotating ring when the elements pass by an array of actuators. For eight-dot Braille, only 12 actuators are needed, each of them capable of turning the cuboid for one line of dots by 90° (Figure 2).

Figure 2: Patent drawing of the BrailleRing concept [14].

Figure 2:

Patent drawing of the BrailleRing concept [14].

Some benefits we originally expected from this approach:

  • – The device will be small and, nevertheless, capable of showing long lines of text.

  • – Having text on the inside of a ring instead on the outside of a drum makes it possible to rotate the device without a motor. If users move their reading finger to the right or left on the inside, they only have limited freedom of movement and by moving their finger further, they will automatically push the device forward or backward, resulting in the rotation needed to display the next or previous Braille characters. On the outside of a drum, users would have freedom to move their fingers to the left or right without any barriers and therefore a separate mechanism to move the ring would be necessary.

  • – The cuboids are bigger than and not as fragile as pins in conventional Braille displays. With this structure and far fewer actuators than conventional Braille displays, the device could become more robust.

The development led to prototypes and mock-ups at vastly different levels of complexity, listed here in the chronological order of their creation:

  • – Very simple ring mock-ups with a fixed, short Braille text on the inside to introduce users to the idea of reading Braille inside a ring. Our first potential user gave his approval to the general concept based on these mock-ups. Variations were later presented to participants in all user studies.

  • – A mock-up on a rail, where a tape with Braille text runs through the lower part of the ring when moving the device. This simulates the reading of a full line of Braille text. Again, this mock-up was approved by our first potential user before we started to work on technological details. This mock-up was also used in the first two bigger user studies. Most participants saw potential in the general concept, but at the same time it was still hard for them to imagine how it would feel like to read a full text with several lines.

  • – A (simplified) working prototype on a scale of 1:2.5. Most mechanical parts were produced on an FDM printer. It is operated by an Arduino Mega and uses nine solenoid actuators to move the cuboids. The software supports the generating of different dot patterns that are easy to read by sighted people who do not know Braille and it can translate short words entered on a PC to Braille. Because of its size and being very slow, this prototype is not suited for user tests, but it proves that the general concept works from a technical point of view (Figure 3A).

  • – A first 3D printed mock-up with movable parts in original size, but without any electronics. It was planned to expand this to a fully working prototype, but we found that the 3D printed parts were too rough. So, we only used this to explain the general concept to people interested in the technology.

  • – A more complex Braille tape mock-up, where the tape is running between two coils that are mounted at the top of the mock-up. This simulates reading of longer texts and without a rail (Figure 3B). It was presented since user study 3 (master students’ mock-up studies).

Figure 3: (A) (top): Working prototype in size 1:2.5. (inner diameter: ∼10 cm). (B) (bottom): Tape mock-up with coils.( inner diameter: ∼4 cm).

Figure 3:

(A) (top): Working prototype in size 1:2.5. (inner diameter: ∼10 cm). (B) (bottom): Tape mock-up with coils.( inner diameter: ∼4 cm).

As already said, most participants saw potential in the general concept, but it is hard for them to imagine what it would feel like to read a longer text on such a device. Even the latest tape mock-up can only show a relatively short text. Therefore, the suitability for daily use can only be tested once our working prototype in original size is finished.

The user involvement suggested that displaying long lines of text is not as important as we assumed it to be. Only a few people own refreshable braille displays with more than 40 cells. In our first round of interviews with 11 participants, only one person who introduced himself as a collector owned a bigger display. This is not only because of the costs and the restricted mobility of the device—some people who used displays with 80 characters in the past also complained about the exhausting arm movement necessary to read a full line.

The benefit they see in line-lengths on our BrailleRing would instead be its flexibility. Ideally, users could separately determine for each line where and when they want to move their hand back to read the next line.

Mobility is also a big topic. In all user studies but the one focussed on mock-ups there were participants who told us they would like to use our device when they are riding on a train. We already had mobility in mind with the small size, but using it in the context of a train ride, most users would not have a table available on which to move the device back and forth. Therefore, a motor for rotation will be needed. To keep the ring itself small and lightweight, we plan to add this motor as an add-on in the future. Some students built early, mains-powered prototypes for such add-ons (Figure 4).

Figure 4: Simple ring mock-up on a prototype adapter for automatic rotation (adapter sizes: 190 × 48 × 55 mm).

Figure 4:

Simple ring mock-up on a prototype adapter for automatic rotation (adapter sizes: 190 × 48 × 55 mm).

The biggest benefit blind people see in our concept is robustness. Two people in our first round of interviews remembered paying between 400 and 800 EUR for a service or maintenance on their Braille display. Later interviews revealed similar experiences and this topic also are tightly connected to mobility: A robust device is much better suited to be used outdoors.

This user feedback gave us a new idea: Our BrailleRing could be constructed in a way so that blind users will be able to clean it by themselves and would no longer be forced to pay for expensive maintenance. The cuboids that display the characters are already pivoted without a direct connection to the actuators or other electronics. The device just needs to be built in a way so that the part with the cuboids can be removed easily for cleaning. Of course, making the device robust and easy to clean will still be a long way from the current state, but the user studies suggest that this should be a focus for further development.

Conclusion

The BrailleRing features a unique approach for a refreshable Braille display with a line of text as long as the user wants it to be. Right now, it is a concept tested in various mock-ups and early prototypes. At the time this paper is written, a first fully functional prototype in original size is in production and is planned to connect to popular screen readers in the future to provide full access to digital texts.

To make the greatest benefit, easy device cleaning, a reality, the inner ring with the Braille cuboids needs to be encapsulated from the electronical parts. Once the display is easy-to-clean and more robust against humidity and dust, it should also be ideally suited for new markets, e.g., in developing countries.

Acknowledgment

We want to thank all blind people who participated in our studies as well as the students that helped us to conduct interviews and to build prototypes and mock-ups.

References

1. DIN 32976. Blindenschrift - Anforderungen und Maße. German Institute for Standardization. Search in Google Scholar

2. Tetzlaff, JF. Patent US 4283 178 A, 1981. Search in Google Scholar

3. Chen, X. Patent WO 2016 197 928 A1, 2016. Search in Google Scholar

4. Wilhelm, E, Schwarz, T, Jaworek, G, Voigt, A, Rapp, BE. Towards displaying graphics on a cheap, large-scale braille display: LNCS, Cham: Springer; 2014, vol 8547, https://doi.org/10.1007/978-3-319-08596-8_102. Search in Google Scholar

5. Luning, X, Han, L, Yufei, L, Shen, R, Kunquan, L. Operational durability of a giant ER valve for Braille display. Smart Mater Struct 2017;26:54003. https://doi.org/10.1088/1361-665x/aa64c4. Search in Google Scholar

6. Liu, Z, Luo, Y, Cordero, J, Zhao, N, Shen, Y. Finger-eye: a wearable text reading assistive system for the blind and visually impaired. In: 2016 IEEE international conference on real-time computing and robotics (RCAR); 2016: 123–8. Conference in Angkor Wat, Cambodia. Search in Google Scholar

7. Roberts, J, Slattery, O, Kardos, D, Swope, B. New technology enables many-fold reduction in the cost of refreshable braille displays. In: Proceedings of the fourth international ACM conference on assistive technologies (Assets ’00); 2000: 42–9. Conference in Arlington Virginia US. https://doi.org/10.1145/354324.354335. Search in Google Scholar

8. Shimamura, Y. Patent EP 1 5222 983 A1, 2004. Search in Google Scholar

9. Roberts, JW. Patent US 2002 0045 151 A1, 2002. Search in Google Scholar

10. Litinskij, PY. Patent SU1633445 A1, 1991. Search in Google Scholar

11. Al-Qudsi, M. Patent US 2013 0203 022 A1, 2013. https://doi.org/10.1109/sbec.2013.20. Search in Google Scholar

12. Treml, M. Grobkonzeption und Steuerung eines Brailledisplays mit gruppierten Punkten [Diploma thesis at TU Wien], Vienna: TU Wien; 2016. Search in Google Scholar

13. Busse, D. Konstruktion eines Braille-Displays [Diploma thesis at TU Wien], Vienna: TU Wien; 2016. Search in Google Scholar

14. Treml, M, Zagler, W, Busse, D. Patent at 518 530 A4, 2016. Search in Google Scholar

15. Krammer, C. Untersuchungen zur Optimierung der taktilen Brailledarstellung in einem ringförmigen Braille Display mittels Versuchsmuster [Master thesis at FH Campus Wien], Vienna: FH Campus Wien; 2019. Search in Google Scholar

16. Seemayer, HP. Analyse des dynamischen Leseverhaltens in Blindenschrift als Grundlage für das Interaktionsdesign in einem ringförmigen Braille Display [Master thesis at FH Campus Wien], Vienna: FH Campus Wien; 2019. Search in Google Scholar