Design and Implementation of Smart Cane

for Visually Impaired People

Do Ngoc Hung, Vo Minh-Thanh, Nguyen Minh-Triet, Quoc Luong Huy,

and Viet Trinh Cuong

Abstract
In this paper, a smart cane can help visually impaired people to detect the obstacles around
them is designed and implemented. A PCB board uses a microcontroller to get data from
ultrasonic sensors to determine the distance from the cane to the nearest obstacles. This
distance information will be transferred to the Android smart phone via Bluetooth protocol.
Whenever user is walking to an area where having obstacles around and the distance
between user and obstacle is shorter than a predefined threshold, the phone will display
distance and give warning by speaking that value in centimeter unit while the vibration
motor attached to the cane will vibrate with the intensity in inversely proportional with the
value of the distance.

Keywords
Smart cane Android Ultrasonic sensor Bluetooth protocol

1 Introduction The authors in [4] introduced a prototype of the mobility

According to the data of World Health Organization (WHO),
the number of visually impaired people are estimated to be
about 285 million with 39 million are blind and 246 million
have low vision [1]. The other data of WHO also shows that
there are more than 29 million people in Western Pacific
Region have problem with their eyes [2]. The most difficulty
of the visually impaired people is how to have safe mobility.
They usually rely on assistance of sighted people or using a
cane to help them in navigation and obstacle avoidance.
There are many aided canes using different types of range
finders have been proposed by researchers and manufactur-
ers to assist visually impaired people in mobility. A smart
cane was implemented by using infrared sensor, ultrasonic
sensor, accelerometer sensor and a compass in [3]. The
correct decision ratios reach over 80% for uneven step
detection and over 90% for dangerous obstacles detection.
D.N. Hung (&)
V. Minh-Thanh
N. Minh-Triet
Q.L. Huy

V.T. Cuong
School of Electrical Engineering, International University
Vietnam National University, HCM City, Vietnam
e-mail: dnhung@hcmiu.edu.vn
recognition system which uses microcontroller to get signals
from ultrasonic sensors and Global Positioning System
(GPS) module to detect around objects and determine
position of user. A Kinect cane system was proposed in [5]
in which particular classes of object can be searched. People
use numeric keypad to give instructions which will be pro-
cessed by a computer. The main disadvantage of this system
is inconvenient in usage because of its bulky size. Recently,
N. Agarwal designed an electronic guidance system with
four modules including object detection, traffic signal indi-
cator, SOS, and landmark identification [6]. They used
Arduino Mega 2560 board and ultrasonic sensor for distance
measurement, Global System for Mobile Communications
(GSM) module for Short Message Service (SMS) messages,
Radio Frequency (RF) module for traffic signal indicator and
Radio Frequency Identification (RFID) tags for landmark
detection. Although this system works well in many cases, it
still meets problem in power consumption because of the
large electric current of Arduino board and modules.
In recent years, Android has become a popular operating
system (OS) for mobiles phone and tablet devices. It has
been developed and distributed by Google based on Linux

© Springer Nature Singapore Pte Ltd. 2018 249

T. Vo Van et al. (eds.), 6th International Conference on the Development of Biomedical Engineering
in Vietnam (BME6), IFMBE Proceedings 63, https://doi.org/10.1007/978-981-10-4361-1_41
250
2.6. Unlikely the other popular OSs for mobile phone like
iOS (developed by Apple) or Windows Phone (developed by
Microsoft), the users who own Android devices have variety
of applications because the developers are able to enhance
their creativity in open source environment. Moreover, the
users have many options to buy Android devices because the
open of the Android OS attracts many manufactures to
develop their products run on Android OS. Besides using
Android OS for mobile phone and tablet, many manufac-
turers also use Android OS for the other devices such as
smart watch, glass, smart television or the automobile
entertainment on the car.
In this study, a smart cane is implemented for visually
impaired people to move independently in a safer way than
using traditional cane which just helps these people avoid
obstacles ahead or distance lower than about 1 m. To
announce as soon as existing obstacles around, smart cane
works with smart phone runs on Android OS can be a better
solution for visually impaired people which is shown in
Fig. 1. The objective of this project includes two parts:
hardware implementation and Android application develop-
ment. The Printed Circuit Board (PCB) board must be
compact to be put inside a cane while the power consumption
is low enough to be used for a long duration with an available
battery. The Android application that can get data through
Bluetooth protocol, display them on the screen and speak
aloud the distance in case it exceeds a predefined threshold.
The rest of this paper is divided as follows: Sect. 2
introduces the overview of the system. Section 3 describes
the steps of building Android application for this project.
Section 4 shows the results of the study. Finally, Sect. 5
concludes the paper and gives some future works.
Fig. 1 Visually impaired person
with smart cane
D.N. Hung et al.
2 System Design
The system model of the smart cane is described in Fig. 2.
There are two main parts including hardware design and
Android application development. The heart of the hardware
is an ATmega microcontroller which used to get data from
the ultrasonic sensors, converts it into distance value, and
send them to Android phone via Bluetooth protocol.
Because this cane is designed for using almost all day, the
battery charger is necessary for recharging the battery easily
whenever the energy level is low. Moreover, the amplitude
voltage of Li-Ion battery is 3.7 V but all the components
need 5 V power supply to run, the step-up voltage is needed
to boot-up voltage from 3.7 to 5 V. One vibration motor
attached to the cane is used to express the distance by its
vibration intensity, so the user can feel the distance to
obstacles. The Android application receives the data from
circuit attached to the cane, displays it on the screen, and
speaks out via speaker or headphone if it exceeds a threshold
which can be set by users.
In this study, ultrasonic module HC-SR04 is used to
measure the distance. This sensor uses sonar to determine
distance to an object like bats or dolphins do. It offers
excellent non-contact range detection with high accuracy
and stable readings in an easy-to-use package.
In measurement procedure, the trigger signal of SR04
must be on high level logic at least 10 ls and wait for the
reflected ultrasonic burst. Whenever the sensor detects
obstacles, the level logic on echo pin will be high and delay
for a periodic which proportion to the distance. The timing
diagram is shown in Fig. 3. Distance D is measured by time
duration the Echo on high level logic as follows:
Obstacle Detected
X cenƟmeters
ahead
Design and Implementation of Smart Cane … 251

Fig. 2 System overview

On/Off Switch Ultrasonic Sensor

Atmega
Step-up Voltage Bluetooth Module Smartphone
Microcontroller

Battery Vibration Motor

pulse width ðlSÞ

D¼ ðcmÞ or
58
pulse width ðlSÞ
D¼ ðinchÞ
148
Serial port Bluetooth HC-06 in Fig. 4 permits any
microcontroller with a standard RS232 serial port to connect
with a PC or a Smartphone. It can play a role in master or
slave mode and unable to switch between two roles. It works
through two line RX/TX and can take baud rate from 1200
to 1,382,400. However, the baud rates greater than 115,200
are not recommended to use because the signal interference
causes instability to the system (Fig. 5).
We use rechargeable battery to support convenient and
low cost usage. Therefore, MCP73831, an integrated circuit
with low price and constant output current, is chosen to
charge the battery. This device is highly advanced linear
charge management controllers for use in space-limited,
cost-sensitive applications. It employs a constant
current/constant voltage charge algorithm with selectable
preconditioning and charge termination. The constant volt-
age is fixed and the constant current value is set with one
external resistor.
All of the components need to be supplied with 5 V DC
voltage source. However, Li-Ion battery just provides 3.7 V
so we need to design a step-up voltage circuit to boost from
Fig. 3 HC-SR04 ultrasonic module 3.7 to 5 V to ensure that all the components work well. The

Initiate Echo back

Pulse width corresponding to distance

10 uS TTS to signal pin
(About 150uS-25mS, 38mS if no obstacle)
Signal

Internal
Ultrasonic Transducer will issue eight 40KHz pulse

Fig. 4 HC-SR04 timing diagram

252 D.N. Hung et al.

Start

Set
Threshold

Turn on Bluetooth
Bluetooth N is on?

Scan the Available Devices

Fig. 5 Bluetooth serial port HC-06

MC34063A series is a monolithic control circuit containing Choose the Board

the primary functions required for DC to DC converters.
These devices consist of an internal temperature compen-
sated reference, comparator, controlled duty cycle oscillator
with an active current limit circuit, driver and high current Connect
output switch. This series were specifically designed to be Successfully?
N
incorporated in Step Down and Step Up and Voltage
Inverting applications with a minimum number of external Y
components.
Receive Data
3 Android Application Development

Android Studio which is the most popular Integrated

Development Environment (IDE) for developing Android
applications [7]. In this study we use this IDE to build an
application runs on Bluetooth-supported Android devices.
The programming language is used to write an Android
application is Java. In case we want to embed the source
code written in C/C++, we can use Native Development Kit
(NDK) toolset. Figure 6 shows all the basic steps in flow-
chart of our project. The main function of this application is
to receive data and capture the event when data have been
sent completely, then speaks out that value in centimeter unit
in case it exceeds a predefined threshold. The application
starts from setting the threshold of distance, turning on
Bluetooth, scanning, pairing the other Bluetooth devices and
speaking out the distance value if the received data is less
than the threshold. The function text-to-speech supported by
Google is used in generating sound from measured distance
in centimeters unit.
4 Result
Figure 7 shows the PCB of the prototype. The board has
compact size (135  21 mm) which can be put inside the
cane easily. The ultrasonic sensors are attached outside the
Waiting for Data Sent Complete Transfer
N ?
Y
N
Data< Threshold ?
Y
Speak out
Fig. 6 Android application flowchart
cane to detect the obstacles in the short range. The Android
smartphone is used to connect with the board via Bluetooth
protocol to receive distance and object’s position
information.
After building the Android project successfully, we have
an Android Application package (APK) file which can be
installed in any smartphone or tablet run Android OS. At the
initial screen when application starts, there are two function
buttons: device searching and device connecting. After
pairing successfully with the cane, smartphone receive
Design and Implementation of Smart Cane … 253

Fig. 7 Printed circuit board

Obstacle
Distance

Measuring Tape

Sensor
Smartphone
(PCB Board)
Fig. 9 Working current measurement

Fig. 8 Block diagram for testing

sensor is 0.3 cm, but in this project we set the resolution is
1 cm because that value satisfies the goal of our product.
distance value from sensors to obstacles, then compare dis- The first column shows the distance measured by using the
tance to a predefined threshold to make decision. User can measuring tape and the third column shows the measurement
listen that value by using headphone or built-in speaker of error. The minimum and maximum distances that the cane
smartphone. can detect is 1 and 340 cm, respectively.
To evaluate the performance of our product, a measuring The error increases when the distance increases due to the
tape is used to measure distance as a reference value to angle of the ultrasonic sensor to the obstacle. It is clear that
compare to the distance displayed on the smartphone’s the error is acceptable for our application with only a few
screen. Figure 8 shows the method to test the product. millimeters in the worst case. The circuit uses two Li-Ion
Table 1 displays the result of the testing. The measure- batteries for PCB and vibration motor. We just calculate the
ment was done in several distances to find out the minimum, power consumption of the PCB board when the ultrasonic
maximum distance that the product can detect and the sensor and Bluetooth module are working. The value of the
measured errors. The minimum resolution of the ultrasonic main current flows through the board is about 310 mA as in
Fig. 9. In this test, the battery has capacity 2200 mAh, so it
can keep the product works on approximately 7 h
Table 1 Measurement results continuously.
Distance (cm) Error (mm)
1 0
5 Conclusions
5 0
This paper describes the design and operation of a smart
10 0 cane which helps visually impaired people can detect the
30 0 obstacles around by vibration intensity of a motor and
50 0 sound generated by Android smartphone. This cane
100 0 length can be changed flexibly so it fits to anybody.
140 −3 The PCB has compact size makes it can be put inside the
190 3 small cane. The power consumption of the cane is rather
low and we can use battery to supply power to entire
250 3
circuit. Furthermore, a battery charger circuit also be
300 4
embedded into PCB, so users can charge the battery
340 −4 easily in case of low level energy. This cane can be
254 D.N. Hung et al.

improved by embedding some sensors and modules to
support users with fall detection, GPS-based direction,
and web-based aided services.
References
1. World Health Organization (2014) Visual impaired and blindness,
fact sheet no. 282
2. World Health Organization (2010) Global data on visual
impairments
3. Niitsu Y, Taniguchi T, Kawashima K (2014) Detection and
notification of dangerous obstacles and places for visually impaired
person using a smart cane. ICMU, Singapore, pp 68–69
4. Abdullah S, Noor NM, Ghazali MZ (2014) Mobility recognition
system for the visually impaired. ISTT, Malaysia, pp 362–367
5. Takizawa H, Yamaguchi S, Aoyagi M, Ezaki N, Mizuno S (2013)
Kinect cane: object recognition for the visually impaired, Poland,
pp 473–478
6. Agarwal N, Iyer A, Naidu S, Rodrigues S (2015) Electronic
Guidance System for the visually impaired—a framework, Mumbai,
pp 1–5
7. Drongelen MV (2015) Android studio cookbook. PACKT Publish-
ing, Birmingham