AndroRC: An Android Remote Control Car Unit for

Search Missions
Yuxin Jing, Letian Zhang, Irwin Arce, Aydin Farajidavar
Integrated Medical Systems Laboratory, Dept. of Electrical and Computer Engineering
New York Institute of Technology
Old Westbury, NY, United States of America
[email protected], [email protected], [email protected], [email protected]

Abstract— The AndroRC is a remote control car (RC) unit be able to navigate deep into rubbles to search for possible
controlled by a smartphone running on an Android application. survivors and to transfer critical data and video to the rescuers.
The car is meant to be used in search missions in the occurrence Information obtained through robots can help the rescuers to
of natural disasters. It is developed to autonomously avoid obtain a quick overview of the affected areas. However, the
obstacles that are not visible to the user driver. The RC unit is robots used in the search missions are often expensive and not
developed based on a Tamiya 70112 Buggy car chassis set with an accessible in most of the undeveloped countries.
extra servo motor added to provide the left and right directions.
The RC is equipped with an ultrasonic distance sensor, a camera, The mobile telecommunications and the internet have highly
a Bluetooth receiver, a Wi-Fi transmitter, two 9-V batteries and penetrated in the daily living of people around the world. The
two Arduino microcontroller boards (UNO and MEGA). The mobile telephony technology is even highly penetrated in
Arduino MEGA controls the propulsion and direction, while the undeveloped counties. It is estimated that in 2014, more than
UNO processes the information received from the distance sensor 4.55 billion people use a mobile phone worldwide. More than a
to stop the RC at a pre-defined distance. third of this population is using a smartphone. As the result the
The Android application uses the embedded orientation market for developing new applications for these smartphones
sensor on the smartphone to determine the four directions is booming [3]. Smartphones are also equipped with inertia and
(forward, backward, left and right) intended by the user; hence, orientation sensors and different communication technologies
rotating the smartphone to different directions results in to the (Wi-Fi, Bluetooth, and cellular) that make them suitable hand-
corresponding propulsion of the RC unit. The control commands held platforms for a variety of novel applications [4].
are transmitted to the RC unit through the Bluetooth
communication. The Android application also receives (via Wi- In this paper, we are presenting an inexpensive remote
Fi) and displays the information from the camera in real-time. control car (RC) platform that can be controlled via a
The AndroRC was characterized and examined on bench-top smartphone. All of the components of the RC platform are
settings. commercially available through a variety of vendors. An
Android application has been developed that uses the
Keywords- Search missions, Remote control car, Android orientation sensor embedded in the smartphone to translate the
devices rotations of the smartphone into various propulsion commands
for the RC unit. The commands are wirelessly transmitted to
the RC unit through a Bluetooth communication. Furthermore,
I. INTRODUCTION
the RC unit streams live videos to the smartphone via Wi-Fi
More than 2 million people were killed in ‘natural’ disasters communication. Fig. 1 illustrates the block diagram of the
worldwide between 1974 and 2003 [1]. Natural disasters such system.
as earthquakes, tsunamis, hurricanes frequently catch us by
surprise and cause large-scale damage to buildings. As a result II. RELATED WORKS
the search and rescue operations happen in the field of
collapsed buildings. The use of heavy machinery is prohibited Similar concepts to this idea have already been developed. A
because they add extra weight on the rubble and destabilize the team from the University of Pretoria in South Africa designed a
structures, which in turn risks the lives of survivors buried in wireless remote controlled car using the internet as a medium
the rubble. Rescue crew often use trained search dogs, cameras [5]. The wireless car was controlled over the internet and it
and listening gadgets to search for victims from the ground streamed live video to the user. This system is bulky and has
surface. Although search dogs are effective in finding human been implemented using computers. Therefore, it cannot easily
under the rubble, they are unable to provide a general maneuver under the rubbles.
description about the environment that the victims are located Police and military forces already have remote controlled
[2]. units in which they use as bomb disposals in certain situations.
In the recent years, robots controlled wirelessly have been The only purpose though is to remove the explosives. These
utilized in the search and rescue missions. These robots should units have very precise robotic arms and stream high quality
 mounted on top of the MEGA board that receives the direction
Distance Arduino commands from the Android device. The Bluetooth operates at
Sensor UNO 5 V and communicates through the 2.4 GHz ISM band. At the
beginning, when the car unit is powered up, the Bluetooth
Bluetooth
receiver (slave) pairs with the master (transmitter) on the
Android device. The servo motor used for controlling direction
Bluetooth Arduino is the SG90, which only weighs 9g and operates at 5 V. The
Receiver MEGA SG90 has a directional turning speed of 0.12 seconds over 60
degrees with no load.
A second processing unit, Arduino UNO board [9], which is
Android based on the ATmega 328, is mounted vertically on the
Servo and
Application wooden chassis next to the MEGA board. The Arduino UNO
DC Motors
board is connected to a distance sensor (ultrasonic ranging
Wi-Fi detector HC-SR04) that is mounted in the front nose of the RC
unit [10]. The microcontroller on the UNO board generates 8
WiFi
Camera
pulses with 1 μs width, to trigger an ultrasonic pulse
Router transmitted from the distance sensor. The sensor listens for
echo. When an echo is received the sensor sends a pulse to the
Fig. 1. The block diagram of the AndroRC is shown. microcontroller. The microcontroller uses the time between
triggering the ultrasound pulse and receiving the echo to
videos in real-time. However, they are expensive and bulky, calculate the distance. A 9-V battery powers up the UNO board
hence, not suitable for search missions. and the distance sensor. The distance sensor provides 2-400 cm
A low-cost quadrocopter with similar concept to the non-contact measurement function and the accuracy can reach
proposed AndroRC unit was designed at the University of up to 3 mm.
Tübingen [6]. The team designed an Unmanned Aerial Vehicle A camera (VGA, 640 × 480) has been mounted on the front
(UAV) that can be controlled remotely using an iPad. They nose to provide the front view of the RC car. The camera is
also have programmed a second quadrocopter that connected to a router device, which is mounted on the back of
automatically follows the lead quadrocopter tracking patterns the RC unit. The videos taken by the camera are fed into the
of emitted infrared lights. The proposed UAV can be deployed router which in turn streamed through the Wi-Fi to the Android
to a variety of missions; however, it is not suited to penetrate device. The videos are displayed on the Android application in
under the rubbles to find survivors. real-time so that the user can make appropriate decisions. The
Arduino UNO board powers up both the Wi-Fi router and the
III. SYSTEM OVERVIEW camera (5 V). Figs. 2 and 3 show the top and front views of the
The system is composed of a remote control car (RC) unit RC unit with all the hardware, respectively.
and an Android smartphone running on a custom-made
application. In the following we have described the RC unit
and the Android application that collectively makes the
AndroRC system.

A. The RC Unit
The AndroRC is based on Tamiya 70112 Buggy Car Chassis
set [7], which is a commercially-available car set, with
adjustable speed settings for high or low outputs. The wooden
chassis was modified to make the unit longer to accommodate
more hardware on top of the chassis. In addition, the front nose
of the chassis was modified to fit an extra servo motor that
rotates the front wheels to the left and right directions.
The Arduino MEGA board [8], which includes an
ATMega2560 microcontroller, is secured in the center of the
wooden chassis. The Arduino MEGA board includes 54 digital
input/output pins (of which 14 can be used as PWM outputs),
16 analog inputs, 4 UARTs (hardware serial ports), a 16 MHz
crystal oscillator, a USB connection, and recommended input
voltage of 7-12 V. The Arduino MEGA board generates the
control pulses for the forward/backward and left/right
propulsions through the dc motor and servo motor installed in
Fig. 2. The top view of the AndroRC shows the camera, Bluetooth, 9-V
the back and in the front of the chassis, respectively. The batteries, Arduino boards, and Wi-Fi router.
MEGA board operates on 5 V. A Bluetooth receiver (BT2S) is
 MEGA UNO
start start

Pairing with the

Ultrasonic
smartphone
sensor
Bluetooth by MEGA

Receiving data Calculating the

from the distance data
smartphone with UNO

Decoding and
Is distance
translating the
shorter than
Fig. 3. The front view of the AndroRC shows the camera, ultrasonic sensor, instructions sent by
20 cm?
Bluetooth, servo motor, Arduino UNO board, and Wi-Fi router. the smartphone

No

Fig. 4 depicts the flow charts that the two microcontrollers

(on the UNO and MEGA boards) follow to execute the Control the DC
defined specifications. If the distance calculated by the motor
ATmega 328 microcontroller on the UNO board is shorter
than 20 cm, an interrupt command will be sent to the
ATMega2560 microcontroller on the MEGA board to shut
down the DC motor to avoid possible collisions. If the Control the DC Shut down the Yes
calculated distance is longer than 20 cm, no interrupt and servo motors DC motor
command will be sent to the MEGA board and the motors Fig. 4. Flow chart of the hardware processors, Arduino MEGA and UNO
continue their routine flow chart (MEGA’s flow chart on the boards.
left side).
B. The Application for Android Smartphone IV. SYSTEM CHARACHTERIZATION AND EXPERIMENTS
An application was developed that can be installed on an The RC unit weighs 850 g and measures 23 × 11 × 9 cm3.
Android smartphone to control the RC unit and receive the With no extra load the maximum speed of the RC unit is
videos. Fig. 5 shows the main blocks of the application. The approximately 50 cm/s. It can carry up to 600 g extra load.
videos received through the Wi-Fi will be captured by the The RC unit is powered up by two 9 V batteries; one for each
InputStream.java and will be displayed on the smartphone’s Arduino board. The current consumption from each battery
screen using the MjpegView.java. A screenshot of the videos was approximately measured as 0.70 A in full load. The
received by application is shown in Fig. 6. There are two details of the current consumption for the MEGA and UNO
objects in front of the RC unit in this figure to provide a view boards are shown in Tables I and II, respectively.
on the distance.
The application also calls the information of the orientation Socketclient.java to transmit data
sensor –embedded in the smartphone– to obtain the angles to the BT
between the phone and a default coordinate defined by the
Android sensor manager. The application then uses pre-defined
thresholds to translate the orientation data into propulsion
commands. These commands are forward, backward, left, and
Main.java
right. For instance, when the angle between the phone and the InputStream.java
Calculates the angle and issues
Y axis breaches -10º, the application sends a "1" to the Arduino From the wi-fi camera
command
MEGA board, which then translates into "forward” propulsion.
The application is designed to be intuitive. Fig. 7 shows the
direction of the propulsion commands based on the orientation
of the smartphone.
MjpegView.java for display

Fig. 5. The main blocks of the Android application

 Z (Yaw), not used

Y (Roll) for forward and

backward commands

Objects
X (Pitch) for left and
right commands

Fig. 6. A screenshot of the video streamed to the application from the RC

unit. Two objects are in front of the RC unit. Fig. 7. The translation of the orientations to different propulsion commands

The maximum range of the Bluetooth and Wi-Fi

communications between the RC unit and the Android device V. CONCLUSIONS AND FUTURE WORKS
were measured as 10 m and 20 m, respectively in open space. We have designed and fabricated a RC unit that can be
The autonomous stopping distance of the RC unit was wirelessly controlled via an Android device (such as a
measured between 1 cm to 10 cm from an obstacle, when set smartphone or a tablet). The system is meant for search
at 20 cm; hence the RC unit did not collide with the obstacle. missions under the rubbles in the occurrence of natural
disasters. The RC unit is small enough to maneuver and pass
through tight spaces. The car is light enough not to add extra
TABLE I. CURRENT CONSUMPTION OF THE ARDUINO MEGA BOARD AT 9 V weight on the rubbles and risk the lives of possible survivors.
Arduino MEGA Power and Voltage The components of the RC unit can be acquired from
Part(s) Current (A)
electronic vendors at the cost under $300. Considering the fact
that the smartphones are abundantly available in most of the
Arduino MEGA 2560 0.04 societies and installation of an application does not impose
MEGA and motor driver 0.10 extra cost, we conclude that the AndroRC is an economical
tool for search missions in the occurrence of natural disasters.
MEGA, motor driver, and BT 0.12 The current measurements of the AndroRC showed that the
MEGA, motor driver, BT, and
0.48 - 0.53
power consumption of the system at full load is relatively
DC motor high. However, knowing which components are consuming
MEGA, motor driver, BT, and
servo motor
0.50 most of the power, we plan to modify the Android application
MEGA, motor driver, BT, DC to accommodate modular control of each component. For
0.72
and servo motors instance, we will turn off the distance sensor through the
Android application, while the camera shows wide open field
with no obstacle. The current consumption test also revealed
that the 9 V batteries have to be substituted with more
TABLE II. CURRENT CONSUMPTION OF THE ARDUINO UNO BOARD AT 9V powerful batteries. This is feasible without much modification,
Arduino UNO Power and Voltage since the AndroRC can easily carry more weights.
In the future, we are planning to add more distance sensors
Part(s) Current (A)
around the RC unit and replace the microcontroller on the
Arduino UNO 0.04 UNO board with a field programmable gate array (FPGA).
UNO and Wi-Fi camera 0.20 - 0.24 This will potentially increase the processing power and speed,
and will further decrease the power consumption. These
UNO and distance sensor 0.50 additional sensors will also help the AndroRC to behave as an
UNO, Wi-Fi camera, and autonomous agent, when the wireless communication with the
0.65 - 0.70
distance sensor Android device is broken.
 REFERENCES

[1] D. Guha-Sapir, D. Hargitt and P. Hoyois,“Thirty Year of Natural

Disasters 1974-2003: The Numbers”, UCL Press, 2004.
[2] A. Ko and H. Y. K. Lau, “Robot Assisted Emergency Search and Rescue
System With a Wireless Sensor Network,” Int. J. of Adv. Sci and Tech.,
Vol. 3, pp. 69-78, Feb. 2009.
[3] E-Marketer. (Jan. 2016). Smartphone Users Worldwide Will Total 1.75
Billion in 2014 [Online]. Available: http://www.emarketer.com
/Article/Smartphone-Users-Worldwide-Will-Total-175-Billion2014/
1010536#qgr DZGBFZdMOv7Bo.99]
[4] B. G. Chun, P. Maniatis, “Augmented Smartphone Applications
Through Clone Cloud Execution,” Conf. Proc. Hot Topics in Operating
Systems (HotOS’09), May 2009.
[5] A. J. Lubbe and P. Kluge, “Development and Testing of a Wireless
Controlled Car Using the Internet as Communication Medium” South
African Journal of Industrial Engineering, vol. 19, no. 1, pp. 137-147,
May 2008.
[6] K. E. Wenzel, A. Masselli and A. Zell, “Visual Tracking and Following
of a Quadrocopter by Another Quadrocopter” Conf. Proc. IEEE Intl.
Rob. Syst. (IROS), pp. 4993-4998, Oct. 2012.
[7] Tamiya USA, Buggy Car Chassis Set [Online], Available:
https://www.tamiyausa.com/items/geniuseries-educational-kits-50/
educational-construction-38000/buggy-car-chassis-set-70112
[8] Arduino, Arduino Mega 2560 [Online], Available:
http://arduino.cc/en/Main/arduinoBoardMega2560
[9] Arduino, Arduino UNO [Online], Available:
http://arduino.cc/en/Main/ArduinoBoardUno
[10] Saint Smart, Ultrasonic Ranging Detector Mod HC-SR04 Distance
Sensor [Online], Available: http://www.sainsmart.com/ultrasonic-
ranging-detector-mod-hc-sr04-distance-sensor.html