Vibrations induced in wheeled equipment by curved protrusions on guide blocks for the visually impaired

ABSTRACT The number of visually impaired people continues to increase as the global population grows and ages, making the provision of textured guide blocks on the floor surfaces of public spaces increasingly critical. However, these blocks can also be obstacles for people with walking difficulties, wheelchairs, luggage, strollers, and other wheeled equipment. This study experimentally measured the vibration levels generated when passing a wheeled bogie over various convex curved guide block protrusions. The curved shape characteristics that induced less vibration in wheeled equipment while still effectively guiding the visually impaired were subsequently identified. The results indicated that curved shapes with widths greater than 13.5 mm and heights less than 5.0 mm exhibited reduced maximum vibration levels compared to the Japanese Industrial Standard (JIS) guide block. Even with a width and height matching those of the JIS block (27.0 mm and 5.0 mm, respectively), the curved shape induced less vibration. However, the maximum vibration level induced by the curved shape was greater than that induced by the JIS block when the height of the former was greater than 5.0 mm, regardless of width. Finally, the faster the bogie passing speed, the greater the maximum vibration level following an approximately linear relationship.


Introduction
The World Health Organization (2022) has reported that at least 2.2 billion people worldwide suffer from myopia or hyperopia.Bourne et al. (2017) noted that although the prevalence of blindness and visual impairment continues to decline, the number of people affected by visual impairment is substantially increasing owing to the growth and aging of the global population.The aging of the population is also expected to increase the number of people who use canes, assistive shopping carts, and wheelchairs owing to their declining physical functions.Indeed, the Cabinet Office of the Government of Japan (2022) reported that approximately 30% of the Japanese population is 65 years or older (the aging rate), and that this percentage is expected to increase in the future.
According to the Japan Federation of the Visually Impaired (2014), guide blocks for the visually impaired are blocks (or plates) provided on the ground or floor that have protrusions on their top surfaces that help to safely guide the visually impaired through the recognition of the tactile sensation on the soles of their feet.These guide blocks represent critical equipment enabling the visually impaired to move independently and safely to their destinations.Though guide blocks such as the JIS T 9251 are inexpensive and easy to install, they have been shown to act as barriers or trip hazards for pedestrians, present difficulties for baby strollers and wheeled suitcases on sidewalks, and hinder the movement of stretchers and medical equipment in welfare facilities and hospitals (Japan Federation of the Visually Impaired 2014; Yamada and Yamazaki 2019).Furthermore, large vibrations can be uncomfortable for wheelchair users and even potentially damage mobile electronic equipment in casters.Therefore, an effective convex protrusion shape for guiding the visually impaired is required that induces less vibration in wheeled equipment than the prismatic projections of JIS blocks.
This study accordingly investigated a curved guide block protrusion geometry to provide a more suitable traveling surface for wheeled equipment by investigating the vibrations generated when a wheeled device passed over a series of fabricated curved protrusion shapes.By analyzing the vibration levels and evaluating the vibration characteristics corresponding to the different curved protrusion surfaces, the most desirable protrusion shapes were identified considering CONTACT Takashi Yamada takashi.yamada0804@gmail.comDepartment of Human Factors Engineering and Environmental Design, Faculty of Biology-Oriented Science and Technology, Kindai University, 930 Nishimitani, Kinokawa-city, Wakayama 649-6493, Japan the comfort and safety of all users.The results of this study can inform the development of smooth guide blocks that provide subtle unevenness to guide the visually impaired while ensuring the comfortable operation of wheeled equipment.The identified block geometry can thereby reduce the labor costs incurred by requiring receptionists and attendants to guide visually impaired and wheelchair-bound people through facilities and improve the experiences of all users of public spaces.

Literature review
There have been several published reviews of studies on guiding the visually impaired.Fernandes et al. (2017) reviewed the assistive technologies proposed by researchers in recent years to address mobility limitations owing to visual impairment.They reported that the visually impaired often require not only information describing their location but also associations between their current location and the features present in the surrounding environment.El-Taher et al. (2021) reported that a variety of assistive devices have been studied and developed over the past decade to help the visually impaired move more safely and independently.
Specific findings have been reported regarding the obstacles and problems faced by the visually impaired when moving: Pigeon et al. (2019) stated that walking without vision imposes a cognitive load, and the more complex the environment, the greater this load; Riazi et al. (2016) suggested that as sidewalks can be among the most dangerous places for the visually impaired, proper urban maintenance is beneficial; Park and Chowdhury (2018) reported that the primary barriers hindering the movement of the visually impaired include inadequate information presentation and obstacles on sidewalks; and Low et al. (2020) evaluated the travel experiences of visually impaired individuals in public transportation, reporting limited access to information, inconsistent infrastructure, and lack of staff support as their primary concerns.Furthermore, Tokuda (2001) reported on specific barriers encountered by people with mobility difficulties, particularly the visually impaired and wheelchair users, highlighting sidewalk steps, obstacles on sidewalks, uneven sidewalks, steep slopes, and improperly laid guide blocks.These results all illustrate the need for pedestrian space improvement to accommodate a wide range of mobilities.
As a result, various studies have investigated the installation of guide blocks to help the visually impaired navigate pedestrian spaces.Tanaka and Iwata (1997) stated that differences in facility planning and philosophies of guide block installation can have unintended effects on users.Indeed, Mizuno et al. (2008) reported that warning blocks (dotted blocks) were often not installed at the intersections of guide blocks (linear blocks) or were improperly installed such that they impeded the movement of people in wheelchairs.Pembuain, Priyanto, and Suparma (2020) stated that the primary problems associated with existing guide block installations included poor connectivity, unsafe routes, inconsistent installation, and sidewalk facilities that were unfriendly to the visually impaired.In addition, Padzi, Ibrahim, and Karim (2013) stated that the arrangement of guide blocks and the user's level of experience were frequently responsible for the deviation of the visually impaired from guide block pathways.
Other studies (Lu, Siu, and Xu 2008;Øvstedal, Lindland, and Lid 2005) investigated the practices and guidelines governing the installation of guide blocks for the visually impaired in different countries.The tactile walking surface indicator (TWSI), which was invented in Japan in 1965, was published as ISO 23599 in March 2012 to facilitate the independent mobility of the visually impaired (Fujisawa and Sueda 2012).The publication of an international standard for TWSIs is expected to promote the global use of guide blocks and contribute to the safety and convenience of the visually impaired in Japan (Ohno, Mizukami, and Tauchi 2014).Ma et al. (2021) confirmed that guide blocks for the visually impaired can help them remain within straight crosswalks, avoid changes of direction, shorten crossing times, and improve the regularity and symmetry of their gait.However, various criticisms of TSWI design effectiveness have been frequently aired in the decades since they were widely implemented (Lu, Siu, and Xu 2008).
Studies have accordingly evaluated the effectiveness of different guide block shapes by conducting walking experiments using visually impaired subjects (Courtney and Chow 2000;Øvstedal, Lid, and Lindland 2005;Ranavolo et al. 2011).Suzuki et al. (2001) reported that the height of the subject, their degree of visual impairment, and differences in test simulations affected the stride length and walking speed of visually impaired people when walking on guide blocks in everyday spaces.Pluijter et al. (2015) stated that future guide block designs should focus on providing directional cues and a more stable surface.
Indeed, findings have been reported concerning the tactile and perceptual information required to guide the visually impaired.Takamiya and Hamada (1998) found that the visually impaired confirm their location when walking using information such as the transient sounds and smells of nearby stores, as well as constants such as roadside facilities and the shape of the sidewalk; however, the visually impaired could become confused if there were many audio guidance points.Parkin and Smithies (2012) reported that the visually impaired can identify different types of surfaces and boundaries and that shared space planning should maintain safe areas for pedestrians.In addition, Rey-Galindo et al. (2020) stated that hearing, touch, and smell are critical sources of information for the visually impaired, demonstrating the need for a signal design that considers these senses.Thus, a contrasting and rich physical environment should be provided that includes tactile surfaces, color contrast, enhanced sounds, and other sensory cues.In this vein, various sensor-based navigation solutions for the visually impaired have been proposed in recent years (Kuriakose, Shrestha, and Sandnes 2020).
Using a systematic study conducted from 1996 to 2000, Mitani et al. (2007) measured the ease with which the visually impaired could detect and recognize guide blocks using a white cane.The results demonstrated that the white cane can easily detect JIS blocks with a ridge height of 5 mm.Indeed, Yanagihara, Kuwahata, and Hara (2013) reported that the rate of detecting projecting shapes using a white cane was 100% in the straight-ahead direction, even with a vertical projection as small as 1 mm.Fujinami et al. (2005) found that a typical guide block width of 300 mm was effective.Takeda et al. (2006) found that linear guide blocks exhibited a higher directional function in a perpendicular arrangement than in a parallel arrangement.In addition, research has investigated the use of orientation blocks attached to guide blocks near crosswalks (Inagaki et al. 2017), the depth of warning blocks suitable for recognition with a white cane (Ståhl et al. 2010), the design and development of guide blocks for navigation in the home environment (Vijaya Prakash and Taduri 2019), and the materials and corresponding recognition of guide blocks for the visually impaired (da Silva et al. 2015;Giordano et al. 2012;Kobayashi et al. 2008;Lee et al. 2016).
Other studies have reported findings concerning the effects of passing over guide blocks.Kobayashi et al. (2003) reported that when clear-sighted people walk on a guide block, they perform certain movements to avoid falling.Encounters with guide blocks have been found to cause the elderly to alter their walking motion (Kobayashi et al. 2006;Thies et al. 2011); there are even indications that the elderly are inconvenienced when walking over guiding blocks (Ormerod et al. 2015).Kobayashi et al. (2005) further stated that a newly designed guide block for the visually impaired, equipped with an indented tactile surface, was effective in reducing extra movement during walking.Furthermore, Mitani et al. (2006) investigated riding comfort when using a wheelbarrow or wheelchair on the Japanese Industrial Standard (JIS) and other guide block geometries, and Bentzen et al. (2020) studied the effects of crossing sidewalk guide blocks installed perpendicular and parallel to the direction of travel on people using various mobility aids.Similarly, Okamura (2008) evaluated wheelchair riding comfort on a tiled surface using vibration levels to clarify the relationship between the tile joint width and wheelchair comfort.Finally, Methods have been developed for the quantification of pedestrian (Wang et al. 2021), wind (Wang et al. 2020a), and seismically (Wang et al. 2022) induced vibrations in bridges and buildings; various tuned-mass dampers have been developed for controlling such vibrations that can inform the development of products subjected to many different types of vibration (Wang et al. 2019(Wang et al. , 2020b(Wang et al. , 2020c(Wang et al. , 2023;;Wang, Shi, and Zhou 2022).
Previous studies have gained considerable insight into pedestrian recognition and comfort when encountering guide blocks for the visually impaired.For example, research has been conducted to optimize the characteristics of the JIS shape by changing its arrangement and dimensions.Other research has evaluated the effectiveness of guide block shapes by surveying visually impaired subjects to identify the most recognizable shapes.However, no research has been conducted to determine the most desirable guide block shape to minimize the vibration of passing wheelchairs and other wheeled equipment.Though previous studies have investigated the vibration level when a wheelchair travels on a floor surface to evaluate riding comfort, the vibration induced by the protrusions of multiple guide blocks for the visually impaired has not been investigated nor have the vibration levels for different block protrusions been compared.

Materials and methods
The evaluated test specimens comprised 10 types of linear protrusions: nine convex curved shapes and the prismatic JIS shape, with the cross sections detailed in Table 1.All test specimens were made of acrylonitrile butadiene styrene and fabricated using a 3D printer.The floor surface surrounding the test specimens was made of urethane-coated linden plywood.Each test specimen was attached to the floor surface using double-sided tape in an orientation such that the long side of the specimen was perpendicular to the travel direction of the bogie.
The bogie vibration measurement device developed by the Floor Construction Working Group of the Architectural Institute of Japan, YukaKoji WG (2020a, 2020b) is shown in Figure 1; the Floor Construction Working Group describes its operation as follows: "The device consists of a bogie with three wheels that is placed on the floor.The vertical vibration generated in the wheels are measured when the bogie is run at a constant speed by winding a rope with a constant speed motor.The measurement is obtained by an accelerometer mounted in the center of the caster." Based on this description, an experimental bogie with one front wheel and two rear wheels was fabricated from stainless steel as shown in Figure 2 and used to measure the vertical vibration level L V generated when it passed over each test specimen.The determined L V represents a logarithmically expressed value of the measured vibration acceleration with human vibration sensory correction (Onosokki 2002).This study used L V to evaluate the sensation of vibration experienced by people with strollers or luggage or riding in wheelchairs.
The bogie was placed on the floor and both its front and rear wheels were pulled over the specimen at a speed of 200 mm/s, 400 mm/s, or 600 mm/s by a rope wound using a constant-speed motor; the resulting L V was measured every 100 ms by a vibration meter (RION VM-55, RION Co., Ltd 2023) attached to the center of the bogie, as shown in Figure 2. To investigate the significance of the differences among the vibrations induced by the various test specimens using the Kruskal -Wallis test (as presented in Section 4.2), each test was performed 30 times to obtain the average maximum vibration level (L Vavg.max ), defined as the average of the 30 maximum L V values for each specimen type (identified as the larger spike in Figure 3).
The types of vibrations that floor surfaces can induce in wheeled equipment include the constant vibration that occurs when wheels travel over gravel or tiled surfaces or the momentary vibration that occurs when wheels encounter a ledge.The maximum vibration level was used for the evaluation in this study because it best describes the temporary vibration that occurs when wheels pass over guide blocks.In the results, reference to "higher levels" or "greater values" indicate that the referenced vibration was more severe than that of the guide block under comparison, whereas "lower levels" or "lesser values" indicate the opposite.high specimen, and the highest L Vavg.max occurred for the 13.5 mm wide by 7.5 mm high specimen.

Vibration level
Comparing vibrations for the JIS specimen with those of the curved specimen of the same width (27.0 mm) and height (5.0 mm), the L Vavg.max of the former was obviously higher.
Figure 5 depicts the relationships between the width of the curved specimen and L Vavg.max , indicating a negative linear relationship at any given height.For heights of 5.0 and 7.5 mm, the slopes of these relationships are comparable, but the slope is notably steeper for a height of 2.5 mm. Figure 6 shows the relationships between the height of the curved specimen and L Vavg.max , indicating a positive linear relationship at any given width; the wider the specimen, the greater the decrease in L Vavg.max as the height decreased.Together, Figures 5 and 6 demonstrate that L Vavg.max was lower for six specimens with curved protrusion heights of 2.5 mm and 5.0 mm than for the JIS specimen.
Figure 7 illustrates the relationship between speed and L Vavg.max for the 27.0 mm wide by 5.0 mm high curved specimen, which match the width and height of the JIS specimen.For both specimens, the higher the speed, the higher the L Vavg.max following a positive linear relationship.However, the slope of this relationship is steeper for the curved specimen than for the JIS specimen.

Kruskal -Wallis test and multiple comparisons
A series of Kruskal -Wallis tests (ScienceDirect Topics 2023), which are used to determine the significant differences between three or more groups with no correspondence, was conducted to further analyze the differences between the L Vavg.max values for the test specimens.In this study, multiple comparisons were made between the JIS and curved specimens using Dunnett's test (Dunnett 1955) in a non-parametric manner (synonymous with Steel's multiple comparison test).Table 2 presents the resulting comparison of the L Vavg.max values (using the JIS specimen tested at a passing speed of 400 mm/s as the Group 1 basis).In these multiple comparisons, the significant differences were below 1% for all curved shapes except for the 13.5 mm wide by 5.0 mm high specimen.The L Vavg.max values of all 7.5 mm high specimens were greater than those of Group 1.
Table 3 provides multiple comparisons of the L Vavg.max values for the JIS specimen and the 27.0 mm wide by 5.0 mm high curved specimen (using the JIS specimen tested at a passing speed of 400 mm/s as the Group 1 basis).The significant differences were below 1% for all comparisons, and the L Vavg.max values for the curved specimens tested at a passing speed of 600 mm/s were all greater than those for Group 1.The symbols <, >, and NS, represent large, small, and not significant differences, respectively Test statistic: Calculated by the Dunnett test Significance: *< 0.01, **< 0.001 Table 3.Multiple comparisons of L Vavg.max at different speeds for the JIS specimen and the 27.0 mm wide by 5.0 mm high curved specimen (using the JIS specimen at a passing speed of 400 mm/s as the Group 1 basis).Table 4 provides multiple comparisons of the L Vavg.max values (using the 27.0 mm wide by 5.0 mm high curved specimen tested at a passing speed of 400 mm/s as the Group 1 basis).The significant differences were below the 1% level for all comparisons.The L Vavg.max values for 13.5 mm wide by 5.0 mm high curved specimen, all three 7.5 mm high curved specimens, and the JIS specimen were all greater than those for Group 1.The L Vavg.max values for all three 2.5 mm high curved specimens and the 40.5 mm wide by 5.0 mm high curved specimen were less than those for Group 1.
Table 5 provides multiple comparisons of the L Vavg.max values for the JIS specimen and the 27.0 mm wide by 5.0 mm high curved specimen tested at different passing (using the 27.0 mm wide by 5.0 mm high curved specimen tested at a passing speed of 400 mm/s as the Group 1 basis).The significant differences were below 1% for all comparisons, and the L Vavg.max values for all curved specimens tested at a passing speed of 200 mm/s were less than those for Group 1.

Discussion
The results of this study can be applied to develop a projection shape for guide blocks that are able to effectively guide the visually impaired while inducing less vibration in passing wheeled equipment.The test specimens with curved protrusion heights of 5.0 mm or less, excluding the 13.5 mm wide by 5.0 mm high specimen, induced lower average maximum vibration levels than the JIS specimen.The average maximum vibration levels induced by the JIS and 13.5 mm wide by 5.0 mm high curved specimens can be considered comparable, as no significant differences were observed between them.Therefore, the shape that best suppresses the maximum vibration levels compared to the prismatic JIS protrusion is a convex curved protrusion 13.5 mm or more in width and 5.0 mm or less in height.When the height of the curved protrusion was 7.5 mm, the average maximum vibration level was consistently higher than that for the JIS protrusion even when the width was increased to 40.5 mm.Furthermore, the relationship between the passing speed and average maximum vibration level was generally linear, with higher speeds resulting in higher average maximum vibration levels.
Even when its width and height were 27.0 mm and 5.0 mm, respectively -the same as the prismatic JIS protrusion -the curved protrusion created a floor surface that induced less vibration in passing wheeled equipment.This geometry is likely to be acceptable for the guidance of visually impaired walkers as a previous study (Yanagihara, Kuwahata, and Hara 2013) found that the visually impaired can recognize a block with a linear protrusion 27.0 mm wide by 1.0 mm high provided for guidance.Thus, the curved projection shapes analyzed in this study can help to develop guide blocks that are easy for visually impaired persons to follow while providing  a comfortable surface for the users of wheeled equipment.Future work should undertake experiments considering guide block material when evaluating the detection and discrimination abilities of visually impaired test subjects.In addition, the vibrations induced in different wheeled equipment types and configurations should be analyzed in detail.Finally, a survey to evaluate the sensation of passing over the proposed protrusion geometry is required to determine how much more comfortable people using wheeled equipment feel when passing over the proposed guide blocks compared to conventional guide blocks.

Conclusions
This study experimentally evaluated the vibration levels induced by the passage of wheels over different convex linear protrusion specimens representing guide blocks for the visually impaired.First, an experimental bogie device for measuring vibration levels was constructed and various test specimens were fabricated with a 3D printer.Vibration levels were measured as the bogie passed over these test specimens.A series of significant difference tests was subsequently conducted to analyze the differences in the average maximum vibration levels in detail, and the results indicated that curved shapes with widths greater than 13.5 mm and heights less than 5.0 mm exhibited reduced maximum vibration levels compared to the prismatic JIS guide block.However, the maximum vibration level induced by the curved shape was greater than that induced by the JIS block when the height of the former was greater than 5.0 mm, regardless of width.This quantitative evaluation indicated that making the linear projection of the guide block curved, widening its width, and lowering its height can significantly reduce the average maximum vibration level in wheeled equipment.Indeed, the results confirmed that a convex linear projection can have a smaller vibration impact on wheelchairs and other wheeled equipment while still effectively guiding the visually impaired.The findings of this study contribute to the development of guide blocks for the visually impaired that do not interfere with wheeled equipment and can serve as a guideline for future designs that ensure the comfortable use of walking spaces by both visually and non-visually impaired people.Future studies should undertake experiments considering guide block materials when evaluating the detection and discrimination abilities of visually impaired test subjects.In addition, the vibrations induced in different types and configurations of wheeled equipment should be analyzed in detail.

Figure 1 .
Figure 1.Specifications of the bogie vibration measurement device proposed by the floor Construction Working Group of the Architectural Institute of Japan, YukaKoji WG (2020a, 2020b).

Figure 3 Figure 2 .Figure 3 .
Figure 3 shows an example L V measurement.The two spikes in the curve represent the L V values when the

Figure 7 .
Figure 7. Relationship between the bogie passing speed and average maximum vibration level (L Vavg.max ).

Table 1 .
Cross-sectional shapes of the 10 test specimens evaluated in this study.

Table 2 .
Multiple comparisons of L Vavg.max (using the JIS specimen at a passing speed of 400 mm/s as the Group 1 basis).

Table 4 .
Multiple comparisons of L Vavg.max (using the 27.0 mm wide by 5.0 mm high curved specimen at a passing speed of 400 mm/s as the Group 1 basis).

Table 5 .
Multiple comparisons of L Vavg.max at different speeds for the JIS specimen and the 27.0 mm wide by 5.0 mm high curved specimen (using the 27.0 mm wide by 5.0 mm high curved specimen at a passing speed of 400 mm/s as the basis).