Computers and Electrical Engineering 67 (2018) 608–619

Contents lists available at ScienceDirect

Computers and Electrical Engineering

journal homepage: www.elsevier.com/locate/compeleceng

Investigating the effects of vibrotactile feedback on human

performance in navigation tasksR
Minghui Sun a,b, Wenzhao Gu c,d, Limin Wang a,b,∗, Liyan Dong a,b,∗, Qian Qian e
a
College of Computer Science and Technology, Jilin University, Changchun 130022,China
b
Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130022,
China
c
National Space Science Center, Chinese Academy of Sciences, Beijing 100190, China
d
University of Chinese Academy of Scineces, Beijing 100190, China
e
Yunnan Key Laboratory of Computer Technology Applications, Kunming University of Science and Technology, Kunming 650500, China

a r t i c l e i n f o a b s t r a c t

Article history: Vibrotactile feedback which can provide real, comfortable and effective interaction experi-
Received 30 August 2017 ences has become indispensable in human computer interaction. Many researchers work-
Revised 14 January 2018
ing with vibrotactile feedback in navigation scenarios have proposed a variety of different
Accepted 15 January 2018
methods of vibration. However, there is little work that has explored the relationship be-
Available online 1 February 2018
tween navigation speed and vibration feedback both in intensity and modes. In order to
Keywords: find the answer, we implemented a prototype supplying with several vibration feedback
Vibrotactile feedback modes and designed two exploratory studies to evaluate the human performance with dif-
User interface ferent navigation speeds. We found using a hand-held vibration feedback device can re-
Navigation system duce the workload of visual and auditory channels. The results also show that the speed
Human computer interaction of navigation has a significant influence on the rate of vibration mode identification. This
paper contributes to the basic understanding of vibration feedback and offers implications
for the future design of vibration feedback in navigation scenarios.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction

It is a big challenge for people to find an appropriate travel path in a strange environment. Nowadays, users usually
learn about the surroundings with navigation instrument. It is helpful to find destination and the spatial position relation-
ship between the current position and the destination. This method is convenient and the applications of navigation system
become widespread. Most of current navigation systems deliver navigation information with visual feedback or audio feed-
back. However, visual and audio channels are often occupied with the first task (driving the car or walking on the street).
It is dangerous to distract users’ attention from traffic information (traffic lights or foot-passengers). Therefore, vibration
feedback is proposed to supply navigation information in this scenario.
Tactile vibration has been used to provide haptic feedback on touch screen equipment [1–5]. When users touch the tar-
get on the screen (such as buttons, drop-down menus, etc.), the vibration motors of device will produce vibration feedback.
This method can help users determine whether they touch the target object or improve the interactive experience. The

R
Reviews processed and recommended for publication to the Editor-in-Chief by Guest Editor Dr. S. Liu.
∗
Corresponding authors.
E-mail addresses: 42600225@qq.com (L. Wang), smh@jlu.edu.cn (L. Dong).

https://doi.org/10.1016/j.compeleceng.2018.01.017
0045-7906/© 2018 Elsevier Ltd. All rights reserved.
 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619 609

expression of vibration technology is an effective mode of human–computer interaction and information delivery methods
[6]. As an important delivery channel of information, it is very important to improve the shortcomings of visual and audio
feedback in specific conditions. For example, when users are in a noisy environment and public places, the navigation voice
supplied from navigation system and other sounds of surroundings mix together. It is hard for users to receive the infor-
mation clearly. When users walk on the street with navigation instruments, it is dangerous and hard for users to stare at
screen and follow the navigation path for a long time. Even for a short time, the limited navigation information is delivered
to the user. When headphones are used to supply audio feedback, it will limit audio feedback from the surrounding envi-
ronment. A recent report [11] shows that 62% of smartphone users under the age of 30 encountered serious car accident
case when they used mobile phones and ignored traffic information. The much more important fact is 43% of people know
this potential threat but still do that. It is necessary and important to drag user’s attention from screen to the street with
“strong” and “intuitive” feedback. In this case, the use of visual and auditory navigation system has been greatly restricted.
And an auxiliary navigation system can be of great help, such as driving tasks [12–14] or pedestrian navigation tasks [15,16].
Vibration feedback technology is a good choice for the auxiliary navigation system. The vibration feedback technology can
not only help people with normal vision, but also have great help for people who have a disability. This technology can also
be used in many ways, such as driving, walking, and other navigation applications.
We found a navigation path by a hand-held device to delivery navigation path information to their user. We adopt this
kind of form mainly because of the current mobile device and the widespread existence in our daily life, and that can make
this kind of technology have great adaptability. Our navigation system includes two types of feedback: speech (semantic)
feedback and vibration feedback. They are supplied to users and provided path information between two exact locations.
We design and implement a vibration navigation system, which provides navigation path information to users by vibra-
tion feedback in many different situations. Our navigation system is composed of a mobile device with a touch screen and
a hardware device that generates vibration feedback. This device supplies vibration feedback with 4 vibration motors and
they are mounted on the four corners. When the user takes the device, it can generate different vibration modes according
to the different positions of the palm and fingers.
In this paper, we first discuss the research on the information related to the mobile touch screen device and the vi-
bration feedback technology. Secondly, we will describe our prototype system and various vibration modes in this project.
Thirdly, we will conduct two experiments to detect whether a user can accurately distinguish different vibration modes with
different walking speeds. Finally, we summarize the whole paper and give the limitation and future work.

2. Related work

Vibration feedback technology has been used as an important interface for the user to interact with the touch screen
device and it has been familiar by the majority of users [1,3,5,24,25]. In this section, we will review the previous vibration
tactile feedback technology on mobile devices, and focus on the wealth of information that is intended to convey, rather
than a simple vibration. The simple reason why we need vibration is that the feedback is more intuitive, and in many cases,
we know our own operation effect without looking at screen.

2.1. Vibration feedback technology on mobile devices

Today, the touch screen has been widely used in a variety of mobile devices, such as smart phones, tablet PCs and music
players. Most of these devices provide users with vibration feedback through the vibration motor embedded. Fukumoto and
Sugimura used the vibrators in Active Click system [1] and let user know the details of their touch event on a touch screen.
In the Active Click system, it can be used for user to know that they have touch an object on the display, and the way of
vibration feedback can help to complete a variety of tasks, such as information input [3] and list item selection [2].
For one vibration motor, it can provide a variety of different vibration modes by changing the frequency and intensity
of the vibration. Poupyrev et al. [5] shows that different vibration modes can convey different information, and the method
of vibration feedback can be used to improve the speed of selection in the linear list with the increase of 1/5. And the
combination of multiple vibration motor can generate more vibration modes, which can be transmitted to the user with
more information. On the other hand, the mobile device can be attached to the vibration motor, which can be used to
provide vibration feedback to the user, rather than the vibration motor embedded in their own. Brown et al. [7] places
the multiple vibration motor attached on the participants’ arm to obtain the vibration feedback to transmit the calendar
information.
These projects mentioned here have shown that different vibration modes can transmit the information of different tasks
to the user. These different vibration modes can be generated by changing the number of vibration motors, the frequency
and intensity of vibration feedback. The multiple vibration motors placed on the body of the participants will have a greater
use of space. However, the placement of multiple vibration motors is also a significant important problem. Sahami et al.
[8] placed six vibration motors on both sides of the smart phone (three per side) in order to test the accuracy of distin-
guishing the different vibration mode. In their experiments, the results showed that the accuracy of the 8 different vibration
modes can be reached by 70–80%, but they are difficult to distinguish the position of the vibration source at each time when
a vibration is generated, and at the same time, the mean accuracy can only reached 36%.
610 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619

2.2. Existing navigation system

Today, navigation system has been widely applied to the daily life. When driving, there will be a vehicle-mounted nav-
igation to guide for us, and when walking, navigation equipment on the mobile device will become essential. These two
kinds of common navigation instruments have become an indispensable product on our daily travel. Both of them use vi-
sual feedback and auditory feedback to remind the travel path. While using the audio feedback, the navigation instruments
transmit the information of path to the user through the voice broadcast function; However, when visual feedback is ap-
plied, the navigation instrument displays the information of path on the screen for the users to see and obtain navigation
path information. However, both of them have some disadvantages. For example, in terms of visual feedback, users can’t
stare at screen while observe a navigation path information for a long time, the screen can not work for a long time and
power supply system will also be a problem difficult to solve; on the other hand, in terms of auditory feedback, if the user’s
environment is too noisy, the voice broadcast will reduce the efficiency greatly, even be in failure, and if the headphones
are used, the sound of great navigation like vehicles whistle sound will be limited, which makes the traveling at a greater
risk of environment. Anita Meier et al, who have studied the human sensory perception which part is more suitable for
vibration, placed vibration feedback system on the top of the shoes [17] and made a confirmed study. However, travelling
speed was not considered in that study.
The application situation of navigation system described above is relatively simple, such as car navigation systems gen-
erally only apply to the process of driving velocity or position of great rapid changed, navigation instrument on the mobile
device can only have a greater effect on the lower speed of movement. Therefore, the use of them has been limited. How-
ever, tactile vibration feedback can play a role in a noisy environment or a variety of moving speed. In a noisy environment,
the auditory feedback is limited, but vibration feedback can be used to provide the information of travel path which can not
be disturbed by surrounding noise; when at different speeds, the vibration feedback is used to provide information of travel
paths, which can not be limited by the speed of travel, such as driving and walking can use the same navigation instrument.
As a consequence, the vibration feedback can play a role in many different situations.

3. System

The vibration navigation system is a kind of navigation system which is designed to be used in different situations and
to use the vibration feedback to remind the information of navigation path. The difference between the vibration navigation
system and the existing navigation system is that it implements the vibration feedback, and uses the voice broadcast and
vibration feedback to transmit the path information. In next section, we will describe the hardware part of the vibration
feedback and voice broadcast, and the different vibration modes.

3.1. Hardware

Fig. 1 shows the part of hardware of our vibration navigation system. Similarly with SemFeel [9], we welded a circuit
board, the circuit board contains four vibration motors, respectively, in accordance with the upper and lower left and right
four directions. The distance between two vibration motors is greater than 1 cm, because there is a study shows that when
the distance of two vibration source is not greater than 1 cm, the vibration source is generated. It will be difficult to distin-
guish where the vibration source is [10]. The circuit board is connected with a single chip microcomputer and a Bluetooth
module. The single chip microcomputer is used to control the vibration of the motor, the Bluetooth module is used to
receive the navigation path information, and the circuit board is connected with a battery box for power supply.
Voice broadcast uses iFLYTEK speech technology to broadcast. In mobile devices, the information of path is transmitted to
the iFLYTEK speech technology broadcast module. Once system obtains the information of path, it uses the speech broadcast
technology to broadcast and complete auditory feedback function.

3.2. Interactive vibration feedback

Our vibration navigation system includes software and hardware. The software is installed on the mobile device, and the
starting point and the destination of the navigation path are finished by the software. After the input is completed, the path
planning is performed to obtain the information of navigation path which is passed to the hardware to generate vibrations
via Bluetooth technology; In the hardware part, the navigation path information is obtained from the software by using
the Bluetooth technology, and the information is processed to control the multiple vibration motors to generate different
vibration feedback.
In our equipment, we use the number of vibration motor and the order of vibration to generate different vibration modes
instead of considering the frequency and intensity of the vibration of these two factors. In the vibration of each vibration
motor, we set the vibration time of 50 0 ms, 10 0 0 ms and 150 0 ms, because a longer period of time of vibration may be not
practical.
 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619 611

Fig. 1. Hardware prototype with cover.

3.3. Vibration feedback mode

Fig. 2 shows that we designed 11 kinds of vibration modes [9]. These vibration modes are divided into four categories:
single motor vibrations (left, right), two motor vibrations (upper-right, upper-left, right-upper, left-upper, down-upper), three
motor vibrations (right-upper-left, down-upper-left, down-upper-right), four motor vibrations (left-down-right-upper). The
11 kinds of different vibration modes represent different turning when driving or walking. In the following experiments,
each vibration motor is only set one kind of vibration intensity in our equipment, because we mainly want to observe the
accuracy rate of different vibration modes, and the recognition degree of vibration feedback mode in many situations.

4. Experiment

4.1. Experiment 1: recognition rate of vibration direction under different navigation speed

4.1.1. Experimental design

Participants. Twelve people (six males and six females, aged from 18 to 40) were recruited for this experiment. One male
and one female used the left hand to carry out the experiment, others used the right hand and all of them had the experi-
ence of using smart phones and other mobile devices with default vibration function. All of participants were compensated
for their time and effort with 20 Chinese yuan.

Apparatus. In this experiment, participants used the vibration device shown in Fig. 3. We installed the software on the an-
droid smart phone, and the software part was written in Java language, running on the android platform. The hardware part
written in C language ran on the C51 microcontroller. The software and hardware parts were connected by the Bluetooth.

Variables. In this experiment, the independent variable included four vibration positions (up, left, right and down) and 13
different navigation speed. The baseline was stationary or motionless. And the maximum speed of running is 12 km/h. Then
separated to 12 points from 1 km/h to 12 km/h. There were three different vibration times, such as 0.5 s, 1 s and 1.5 s. In
612 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619

Fig. 2. Eleven kinds of vibration modes.

 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619 613

Fig. 3. Vibration direction (U represents upper, L represents left, R represents right, D represents down).

Fig. 4. Walking on a treadmill.

this experiment, the vibration times and navigation speeds were the main variables of our study. Each participant needed
to take part in:

5616 times in total (12 participants × 4 vibration positions × 13 navigation speeds × 3 vibration times × 3 trials)

Procedure. Before beginning experiment, we explained the purpose of experiment and experimental process to the partici-
pants. After the end of the explanation, the participants started to be familiar with the equipment and vibration modes. In
each experiment, equipment was vibrated at least three times, and the interval time was 5 s. After three kinds of vibration
were completed, the participants rated the ease of perception of the vibration on a 7-point Likert scale [19] (the 1 repre-
sents the most difficult to perceive the vibration mode, and the 7 is the easiest to perceive the vibration mode) to record
every time the degree of perception. Meanwhile, the participants made an oral evaluation of the vibration mode, and the
experiment organizers took notes.
In this experiment, the direction of vibration was random. Each direction of the experiment was in the three movements
of motionless, walking and running. As shown in Fig. 4, the participants were asked walking or running on a treadmill. They
can use dominant hand to hold our prototype.
The experiment took approximately 35 min per participant. After the end of all the trials, each of the participants com-
pleted a subjective questionnaire. In the questionnaire, the participants were asked about the interval of time of vibration,
the vibration mode, the system’s evaluation and improvement suggestions and other issues. When the experiment was over,
614 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619

Fig. 5. Recognition rate with different navigation speeds.

the participants were asked if there were suggestions or questions for system. It took around one hour for each participant
to finish the experiment.

4.1.2. Results
We measured the recognition of direction with vibration feedback (shown in Fig. 5). If participants said the right direc-
tion system pointed to, the number of correct recognition was added. The experiment was within subject design and was
counterbalanced to eliminate bias by random experimental conditions. The main effect (F2,22 = 3.413, p = 0.055) from vibra-
tion time on recognition was not clear. The value of p was just on the boundary of 0.05. An ANOVA test showed that there
was significant effect (F12,132 = 432.1, p = 0) from navigation speed on recognition. The correct recognition under motionless
condition was the highest among these three conditions. There was also significant interaction effect (F24,264 = 13.16, p = 0)
from navigation speed × vibration times. It suggests that the mode and intensity of vibration feedback should be adaptively
supplied according to the different navigation speeds.
When the speed was less than or equal to 5 km/h, it was in a state of walking, and when the speed was greater than
5 km/h, it was running. After measuring, when the vibration time was 500 ms and the participants were able to feel the
vibration, the maximum speed could reach 10.5 km/h. At that speed, we felt weak vibration. On the other hand, when the
vibration time was 1500 ms and the participants were able to feel the vibration, the maximum speed could reach 11.2 km/h.
At that time, we also had felt weak. In summary, when the traveling speed is less than 9 km/h, the accuracy of vibration
perception of the four directions could reach more than 85%. When the traveling speed between 10 km/h and 11 km/h, the
accuracy of vibration perception would gradually reduce. When the speed was greater than 11 km/h, the three vibration
time of the current equipment was close to the failure. However, if the vibration time were increased, it will still play a role
in vibration perception.

4.1.3. Discussion
The experimental data and conclusions are basically satisfied with the original idea. In this experiment, the accuracy of
vibration perception can reach more than 90%, and the error of the vibration perception is caused by the following several
points:

1) Vibrator is stuck to the circuit board, so all the vibration equipment is in vibration. Thus, it is a little hard for users to
perceive the vibration correctly. To avoid mistakes, subjects are asked to hold the whole device and they can feel the
vibration positon clearly.
2) In the actual walking process, when the walking speed is fast, but the intensity of vibration is weak, the vibration is
not detected obviously.
3) When the vibration time is too short, vibration perception is not obvious and not easy to distinguish the direction of
vibration.
4) The actual initiation voltage of each vibrator is different, when using the same power to supply, the vibration intensity
is not consistent.

We only need to miniaturize the device and change the time of vibration to enhance the user’s vibration perception. In
addition, we need to further explore the recognition rate of our vibration mode at different speeds in walking process.
 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619 615

Fig. 6. Walking in the real world.

4.2. Experiment 2: recognition rate of vibration mode under simulated path

4.2.1. Experimental design

In experiment 1, we evaluated the human performance of recognizing vibration feedback with different speeds in the
gym. In experiment 2, we try to evaluate vibration feedback in the real world. We used an APP named Baidu map and
programmed with its SDK. A starting point and destination were setup (for example, from university to the nearest bus
station) and participants were asked to following by vibration feedback. They can walk or run three different speeds (slow,
middle, fast).

Participants. Twelve people (six males and six females, aged from 18 to 40) were recruited for this experiment. One male
and one female used the left hand to carry out the experiment, others used the right hand and all of them had the experi-
ence of using smart phones and other mobile devices with default vibration function. All of participants were compensated
for their time and effort with 20 Chinese yuan.

Apparatus. The same as experiment 1.

Variables. In this experiment, the independent variable included eleven vibration modes and three kinds of movement (slow,
middle, fast). According to the study, the average walking speed was 1.25 m/s, we set the average value for our experiments
in the speed of walking [18]. The speed of running would be changed according to different people and different external
factors. We had done a preliminary experiment before the experiment for the navigation speed. We made the average value
of 4.5 m/s as the speed of our running as middle. For fast speed, we suggested them to run as 6.5 m/s. In this experiment,
we measured the perception rate of vibration modes. As shown in Fig. 2, we chose 11 vibration modes to guide participants.
In this experiment, the vibration mode and three different speeds were the main variables of our study. Each participant
needed to take part in:

1188 times in total (12 participants × 11 vibration modes × 3 navigation movements × 3 trials )

Procedure. Before the start of the experiment, we explained purpose of the experiment and the experimental process to
the participants. After the end of the explanation, the participants started to be familiar with the equipment and vibration
modes (see Fig. 3). We randomly selected a simulation path to generate vibration feedback for participants to perceive and
distinguish, the participants rated the ease of perception of the vibration on a 7-point Likert scale [19] (the 1 represents the
most difficult to perceive the vibration mode, and the 7 is the easiest to perceive the vibration mode) to record the degree
of perception every time. Meanwhile, the participants made an oral evaluation of the vibration mode, and the experiment
organizers took notes. As shown in Fig. 6, the participants were asked walking or running in the real world. They can use
dominant hand to hold our prototype.
616 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619

Fig. 7. Recognition rate with different navigation movements.

4.2.2. Results
Different from experiment 1, the participants were asked to evaluate our prototype in the real scenario. We focused on
the evaluation of the whole idea and system. Our hypothesis were that the recognition rate of every vibration mode reached
more than 80%, and it guided the user to walk or run effectively in accordance with the established route at different speeds.
In the pilot study, results showed that when the time interval of the two different vibration modes was much longer than
the vibration interval between two vibrators, the recognition rate of vibration mode could reach 70–85%. However, when the
time interval of the two different vibration modes closed to the vibration interval between two vibrators, the recognition
rate of vibration mode could only reach 50%. In order to avoid this problem, the time interval of two modes was designed
longer than the one of two vibrators in experiment 2.
We measured the recognition of direction under different conditions with vibration feedback (shown in Fig. 6). If par-
ticipants said the right modes, the number of correct recognition was added. The experiment was within subject design
and was counterbalanced to eliminate bias by random experimental conditions. As shown in Fig. 7, the recognition rate
with slow speed is higher than other two speeds. However, an ANOVA test showed that there was no significant ef-
fect (F2,22 = 2.84, p = 0.754) from three different navigation movements on recognition. There was also no significant effect
(F10,110 = 7.82, p = 0.89) from vibration modes on recognition.
From the results of subjective evaluation, most of participants like this system and highly evaluated usability. One of
them said “the system can help me to focus on walking on the street. I can ‘watch my step’ while follow the guidance with
this system”. Another participant said that “I often drive the car but can’t find the destination. So sometime I have to stop
the car on the street and check the navigation application. It’s dangerous and waste a lot of time. It is much better than
audio feedback since family members often like listening music and talking aloud in car. ”

4.2.3. Discussion
Our second experiment showed that the whole design idea was correct, and it can provide guidance for the pedestrian in
a wide variety of traveling speeds. However, there are a number of constraints on the device which can not fully achieve the
desired standards in the specific implementation. The main reason for this situation is that the relationship of size between
the time interval of the two different vibration modes and the vibration interval between two vibrators. When the time
interval of two different vibration modes is shorter or close to the vibration interval between two vibrators in the same
mode, there will be some problems with the distinction between different vibration modes; when the time interval of two
different vibration modes is long or much longer than the vibration interval between two vibrators in the same mode, the
recognition rate of vibration modes is relatively high and it is easy to distinguish different modes of vibration and complete
their respective prompt function. Because of the limitations of device, the recognition rate of the current system is not
satisfactory. But we believe that with the continuous improvement of equipment and technology continues to develop, our
equipment will continue to improve, our system will be better and the proportion of the impact factor of device limitation
will be smaller and smaller.
We believe that we can accomplish the goal of assisting the existing navigation technology or navigation equipment by
improving the identification accuracy of vibration mode, and can provide users with more convenience. In the future work,
we will gradually improve the vibration navigation system (for example, to modify the design of vibration mode), in order
to improve the recognition accuracy of vibration mode. And we will consider recognizing different users [20,21,26] with
different custom vibration feedback [22,27]. Meanwhile, we will also apply the vibration feedback modes to more fields,
giving more physical meaning to the vibration feedback mode, so that it can be applied in other fields, such as the rapid
selection of the touch screen, tracking moving target [23].
 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619 617

5. Conclusion

Vibration feedback effectively increases the application range of the navigation system, especially when the user can not
directly view the touch screen. In this paper, we restore the hand-held vibration feedback device which connected with a
plurality of vibrators embedded to explore the recognition rate of vibration feedback modes under different moving speed.
Our findings showed that navigation speed affected the recognition rate of vibration pattern significantly. This paper con-
tributes to the basic understanding of vibration feedback and offers implications for the future design of vibration feedback
in navigation scenarios. Our two experiments also showed that the accuracy of the user’s identification of the vibration
direction can reach about 90% when the travel speed is less than or equal to 12 km/h. For the 11 vibration modes, the ac-
curacy of identification is not high and only about 70%, but the idea or direction is correct, it just needs to improve the
experimental equipment and experimental steps to improve the identification accuracy of the vibration modes.

6. Limitation and future work

There are two potential limitations of our study. First, the number of strength and models of vibration feedback are
limited, therefore, the results are concluded by this experimental setup. Second, all data is self-reported. Although we asked
12 participants to take part in the experiment, different people may have different feelings and the results may be affected
by subject’s sensitivity. We hope to address these limitations mentioned above in future work.
The experimental results of this paper show that the design of vibration modes in vibration navigation technology needs
to be further improved in order to improve the recognition accuracy of vibration mode. Because the vibration mode is much
higher, the vibration feedback technology can be better to assist users in navigation. In addition, this navigation can be
carried out in a variety of situations to navigate, without considering the changes in the environment (within the carrying
capacity of the device). This system can make people’s sense focus on a single task, as far as possible to avoid the emergence
of a variety of tasks in the same sense. We believe that this vibration navigation system will be used as an auxiliary system
for existing navigation system to give many travelers great convenience.

Acknowledgments

This study has been partially supported by National Key Research and Development Program of China (No.
2016YFB1001304), the State Key Program of National Natural Science Foundation of China (No. 61631010), the Science and
Technology Development Program funded projects of Jilin Province (20150520065JH) and Educational Commission of Jilin
Province (JJKH20170803KJ).

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.compeleceng.
2018.01.017.

References

[1] Fukumoto M, Toshiaki S. Active click: tactile feedback for touch panels. CHI’01 extended abstracts on human factors in computing systems. ACM; 2001.
[2] Hall M, Hoggan E, Brewster S. T-Bars: towards tactile user interfaces for mobile touchscreens. In: Proceedings of the 10th international conference on
Human computer interaction with mobile devices and services. ACM; 2008.
[3] Hoggan E, Brewster SA, Johnston J. Investigating the effectiveness of tactile feedback for mobile touchscreens. In: Proceedings of the SIGCHI conference
on human factors in computing systems. ACM; 2008.
[4] Hoggan E, Anwar S, Brewster S. Mobile multi-actuator tactile displays. Haptic Audio Interact Des 2007:22–33.
[5] Poupyrev I, Maruyama S, Rekimoto J. Ambient touch: designing tactile interfaces for handheld devices. In: Proceedings of the 15th annual ACM sym-
posium on user interface software and technology. ACM; 2002.
[6] Wang G, Zheng J, Fan H. Route guidance for visually impaired based on haptic technology and their spatial cognition. International conference of
pioneering computer scientists, engineers and educators. Singapore: Springer; 2017.
[7] Brown LM, Brewster SA, Purchase HC. Multidimensional tactons for non-visual information presentation in mobile devices. In: Proceedings of the 8th
conference on human–computer interaction with mobile devices and services. ACM; 2006.
[8] Rantala J, Roope R, Jani L, Veikko S, Jukka R, Katri S, Toni P, Arto H. Methods for presenting braille characters on a mobile device with a touchscreen
and tactile feedback. IEEE Trans Haptic 2009;2.1:28–39.
[9] Yatani K, Truong KN. SemFeel: a user interface with semantic tactile feedback for mobile touch-screen devices. In: Proceedings of the 22nd annual
ACM symposium on user interface software and technology. ACM; 2009.
[10] Palmer CLAUDEI, Gardner ESTHERP. Simulation of motion on the skin. IV. Responses of Pacinian corpuscle afferents innervating the primate hand to
stripe patterns on the Optacon. J Neurophysiol 1990;64.1:236–47.
[11] EARS and EYES. Handy im Straßenverkehr: Jüngere unterschätzen das Risiko. Ears and Eyes Creating new Grounds; 2014 https://www.earsandeyes.
com/de/presse/handy- im- strassenverkehr/.
[12] Labiale G. In-car road information: comparisons of auditory and visual presentations. In: Proceedings of the human factors society annual meeting,
vol. 34 (9). Sage CA: Los Angeles, CA: SAGE Publications; 1990.
[13] Liu Y-C. Comparative study of the effects of auditory, visual and multimodality displays on drivers’ performance in advanced traveller information
systems. Ergonomics 2001;44.4:425–42.
[14] Zeichner N, Phoenix P, Miranda S, Laura B, Tim N. Exploring howmobile technologies impact pedestrian safety. NYC Media Lab Research Brief; 2014.
[15] Pielot M, Boll S. Tactile Wayfinder: comparison of tactile waypoint navigation with commercial pedestrian navigation systems. Pervasive Comput
2010:76–93.
618 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619

[16] Rümelin S, Rukzio E, Hardy R. NaviRadar: a novel tactile information display for pedestrian navigation. In: Proceedings of the 24th annual ACM
symposium on user interface software and technology. ACM; 2011.
[17] Meier A, Matthies DJC, Bodo U, Reto W. Exploring vibrotactile feedback on the body and foot for the purpose of pedestrian navigation. In: Proceedings
of the 2nd international workshop on sensor-based activity recognition and interaction. ACM; 2015.
[18] Knoblauch R, Pietrucha M, Nitzburg M. Field studies of pedestrian walking speed and start-up time. Transp Res Rec 1996;1538:27–38.
[19] Dawes J. Do data characteristics change according to the number of scale points used. Int J Market Res 2008;50.1:61–77.
[20] Liu S, Pan Z, Fu W, Cheng X. Fractal generation method based on asymptote family of generalized mandelbrot set and its application. J Nonlinear Sci
Appl 2017;10.3:1148–61.
[21] Liu S, Lu M, Liu C, Pan Z. A novel distance metric: generalized relative entropy. Entropy 2017;19.6:269.
[22] Jiang C, Uchida K, Sawada H. Research and development on portable Braille display using shape memory alloy wires. In: Proc. of 4th international
conference on human system interaction, May 19–21; 2011. p. 318–23.
[23] Pan Z, Liu S, Fu W. A review of visual moving target tracking. Multimedia Tools Appl 2017;76(16):16989–7018.
[24] Sun G, Xue Y, Dong Y, Wang D, Li C. A novel hybrid method for effectively classifying encrypted traffic. In: Proceedings of IEEE globecom, December
7–9, Miami, USA; 2010.
[25] Sun G, Li S, Chen T, Li X, Zhu S. Active learning method for Chinese spam filtering. Int J Performability Eng 2017;13(4):511–18.
[26] Sun G, Li S, Cao Y, Lang F. Cervical cancer diagnosis based on random forest. Int J Performability Eng 2017;13(4):446–57.
[27] Jiang C, Uchida K, Sawada H. Research and development of vision based tactile display system using shape memory alloys. Int J Innovative Comput Inf
Control 2014;10(3):837–50.
 M. Sun et al. / Computers and Electrical Engineering 67 (2018) 608–619 619

Minghui Sun received the Ph.D. degree in computer science from Kochi University of Technology, Japan, in 2011. He is currently an assistant professor in
the college of computer science and technology in Jilin University, China. He is interested in using HCI methods to solve challenging real world computing
problems in many areas, including tactile interface, pen-based interface and tangible interface.

Wenzhao Gu received his undergraduate in Jilin University, China, in 2015. Now he is a master student at University of Chinese Academy of Sciences and
National Space Science Center, CAS. His research interests include human–computer interaction, tactile presentation and Compressive Sensing.

Limin Wang received the Ph.D. degree in computer science from Jilin University in 2005. He is currently a professor in Jilin University, China. His research
interests include probabilistic logic inference and Bayesian network. He has published innovative papers in journals such as Knowledge Based Systems,
Expert System with Applications and Progress in Natural Science.

Liyan Dong received the Ph.D. degree in computer science from Jilin University in 2008. He is currently a professor in Jilin University, China. His research
interests include machine learning and Data mining.

Qian Qian was born in 1981. He got his Ph.D. degree from Kochi University of Technology, Japan, in 2011. Now, he is an associate professor in Kunming
University of Science and Technology, China. His main research interests lies in human information processing, such as visual attention.