See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/224503109

Navigation of mobile robot using Global Positioning System (GPS) and obstacle
avoidance system with commanded loop daisy chaining application method

Conference Paper · April 2009

DOI: 10.1109/CSPA.2009.5069211 · Source: IEEE Xplore

CITATIONS READS

28 9,288

4 authors:

Mohamad hanif Abd Hamid Abdul Adom

Universiti Malaysia Perlis Universiti Malaysia Perlis
3 PUBLICATIONS 30 CITATIONS 181 PUBLICATIONS 2,031 CITATIONS

SEE PROFILE SEE PROFILE

Norasmadi Abdul Rahim Mohd Hafiz Fazalul Rahiman

Universiti Malaysia Perlis Universiti Malaysia Perlis
97 PUBLICATIONS 671 CITATIONS 275 PUBLICATIONS 1,921 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Mohd Hafiz Fazalul Rahiman on 03 May 2020.

The user has requested enhancement of the downloaded file.

2009 5th International Colloquium on Signal Processing & Its Applications (CSPA)

Navigation of Mobile Robot Using Global

Positioning System (GPS) and Obstacle Avoidance
System with Commanded Loop Daisy Chaining
Application Method
MHA Hamid, AH Adom, NA Rahim, MHF Rahiman
School of Mechatronics Engineering, Universiti Malaysia Perlis 01000 Kangar, Perlis, Malaysia
hanif@unimap.edu.my

Abstract- Mobile robot has been a major role to the
application in military, industrial and agricultural
purposes. Mobile robot should navigate through desire
route and avoid the obstacle within the path. Many
researcher come with the solution by using the various
type of control and instrumentation system. The
complexity of mobile robot system can make the system
cost intensive and high risk. Proposed is a mobile robot
equipped with GPS navigation system and obstacle
avoidance system with low cost mobile structure, GPS
module and sonar sensor. The combination GPS and
Sonar will determine the position and obstacle avoidance
for the mobile robot. Mobile robot should navigate
according to waypoint that preset to the GPS module and
sonar sensor detects the obstacle during mobile robot
navigation by triggering the sonar sensor in sequence by
using commanded loop daisy chaining application method.
Mobile Robot can navigate through desired waypoint and
at the same time apply the obstacle avoidance rules
robot for de-mining [3]. All of these applications will save life
and suitable for unexpected war nowadays.
A possible task of Autonomous Mobile Robot is ready to
navigate from point to point that pre-programmed routed to
Autonomous Mobile Robot while avoiding the obstacle. This
function is useful in applications such as an autonomous log-
skidder, security surveillance robot, fire suppression system,
and a terrain-mapping vehicle. The vehicle can accomplish
this task by using sensors to “see” where it is and what is
around it. These sensors vary from close range infrared
sensors to longer ranged high frequency radar and global
positioning satellites. [1]
This project used low cost equipment and instrumentation.
Remote control truck with skid steer application is used as a
test bed for GPS navigation and obstacle avoidance. This
remote control truck is suitable on semi-rugged terrain such a
football field and outdoor application. The system was
designed as a prototype for small vehicle and it can be
implemented to larger vehicle for various type of application.

Keyword- Global Positioning System (GPS), National Marine II. MOBILE ROBOT SPECIFICATION
Electronics Association (NMEA), Autonomous Mobile robot,
Sonar Sensor, Daisy Chaining. Mobile robot for this experiment utilized the use of remote
control (RC) truck with the dimension of 50 cm in length, 25
cm in width and 15 cm in height. The RC truck equipped with
I. INTRODUCTION four independent motor that can actuated individually which
can skid steer to turn the mobile robot in immediate 360
Autonomous Mobile Robot is an automated vehicle that degree turn. Equipped with seal lead acid battery, mobile robot
does not required operation from operator to navigate the increases the capability to move in longer time. Figure 1 show
robot. All the movements are pre determined before the mobile robot structures integrated with GPS and sonar sensor.
operation and the mobile robot will navigate accordingly.
Autonomous mobile robot can be divided into three different
type of area which is ground, air and submarine. The
implementation of autonomous mobile robot such as in
military, agricultural purposes, surveillance and other
application can lower the risk of injury and time consuming.
Many researchers are spending a lot of time and money to
develop an autonomous mobile robot. Applications on
military for bomb defusing give major impact to reduce the
fatal injury. In other area the military can use the autonomous
mobile robot to rescue injured army from the battle field.
Other application for military purposes is autonomous mobile

978-1-4244-4152-5/09/$25.00 ©2009 IEEE 178

2009 5th International Colloquium on Signal Processing & Its Applications (CSPA)
Fig. 1. Mobile Robot Structure
III. MICROCONTROLLER
There are many micro controllers in the market consist of
various type capability from basic input output to high end
micro controller. These various types of micro controller are
purposes made for general application. For this project the
suitable micro controller is Basic stamp BS2 and BS2p. For
BS2 it offer processor speed 20 MHz and it can execute
approximately 4000 instruction per second. RAM size for BS2
is 32 bytes and 2K bytes memory for EEPROM about 500
instructions can be store in flash memory. BS2 also offer 16
input output for interfacing with hardware and 2 dedicated
serial communications. BS2P is an upgrade unit
microcontroller from BS2 that offer processor speed 20 M Hz
and execution speed approximately 12000 instruction per
second. RAM size is 38 bytes and flash memory to store
programming code is 16k bytes roughly about 4000
instruction can be saved. 24 input output and 2 dedicated serial
communication integrated in BS2p module. Voltage
requirement for both BS2 and BS2p vary along 5-12 vdc and
current draw at 5v is 40 mA and 350 µA for inactive situation.
Programming language for both controllers is using PBASIC
code. GPS data acquisition will be store in BS2p by serial
communication and all the movement and obstacle avoidance
come from the BS2. Figure 2 show block diagram of the
mobile robot system.
Fig. 2. Block Diagram for Microcontroller Input and Outputs
978-1-4244-4152-5/09/$25.00 ©2009 IEEE
IV. NAVIGATION
Global Positioning System (GPS) is widely use in
navigation and localization. Nowadays many vehicles are
equipped with GPS for navigation from desire route. From the
GPS user can determine which direction they should follow
with time and other data such as bearing etc. This project will
benefit the use of data transmit from satellite to GPS module.
Autonomous mobile robot for this application will use
Garmin E-trex Vista for GPS module. Many considerations
have been made for this application. The main factor for the
consideration is the 12 channel GPS receiver designed to
operate with L1 frequency and GPS receiver continuously
tracks and uses the signal to compute and update position.
Small sizes make it light and easy to handle and also can be
use in rugged terrain because of water proofing. Due to
minimum power requirement which is 3v it is suitable for
autonomous mobile robot that needs longer power for
localization and navigation. With proper calibration the
accuracy will be ±2 with ±5 degrees extreme northern and
southern latitudes and 1 degree resolution.
The experiment has been conducted and the data that from
the GPS that suitable for navigation and localization is
$GPRMC, $GPRMB, $$GPGGA and HCHDG. GPS data
being utilized in this project include latitude and longitude and
also bearing. Specified waypoint can be pre determine and the
autonomous mobile robot will follow waypoint accordingly.
All the data will be read as ASCII code that need to be parse
to various type of information. Usually the accuracy of the
GPS module can up to 9 meter.
Fig. 3. Mobile robot Navigation using heading and waypoint information.
Mobile robot will navigate as shown in Figure 3 through
desire point according to rule below:
Lets
True Course=Heading = TR
Waypoint Heading = WP
If WP>TR Move RIGHT
If TR>WP Move LEFT
179
2009 5th International Colloquium on Signal Processing & Its Applications (CSPA)
If TR=WP Move FORWARD
Garmin E-Trex Vista will submit string to micro controller
through serial port in form of text file. Microcontroller will
recognize the sentence by identifying the character of
$GPRMC at the beginning of the sentence. The data
description for $GPRMC can be shown as Figure 4. Many
data can be used as a guideline for navigation. For example
latitude and longitude can be parsed respectively data 3 and 5.
This data can be use as an initial coordinate where the mobile
robot to navigate through specified route.
Fig. 4. Data string from Garmin E-Trex Vista
All $GPRMC data must be valid status for new value for each
point before continue for another string.
Figure 5 show graph for latitude value from $GPGGA
which has been taken before the navigation has been
conducted to see the accuracy of the GPS module. Four
hundred data has been taken. Within 300 data all the data start
with the same value which indicates the value for latitude is
ready to be used for navigation. Figure 6 show graph for
between longitude and latitude.
Fig. 5. Graph Latitude versus Number of data
978-1-4244-4152-5/09/$25.00 ©2009 IEEE
Fig. 6. Graph Longitude versus Latitude
V. OBSTACLE AVOIDANCE
Along the navigation from the first waypoint to final
waypoint sonar sensor will detect the obstacle within range.
Basic principle of movement of mobile robot is avoid the
obstacle and after successfully avoid the particular obstacle
mobile robot will correct the waypoint from the current
position. This paper focused on the LV-MAX Sonar for
obstacle avoidance system. The selection of sonar sensor is
base on the specification of the LV-MAX Sonar. The
advantages of the LV-MAX Sonar are very low cost sonar
sensor with reliable and stable range data and also very fast
measurement cycle.
On the mobile robot four sonar sensors implemented in
front of mobile robot for obstacle avoidance system. For each
left and right the position of the sensor are 20 cm in between
with 45 angle and two sonar sensor in the middle of left and
right sonar sensor with 10 cm between both sensor.
The left and right sensor will detect the obstacle on each
side and the centre sensor will detect the opposite obstacle of
mobile robot. The system did not include back sensor because
mobile robot only avoid obstacle by moving left and right
only. From this experiment output data from the sonar sensor
using pulse width output. Pulse data can be converted to
centimeter or inches from equation (1) and (2).
The main problem of sonar sensor is the crosstalk between
those sensors. For this system it will have 4 main sonar
sensors in front of the autonomous mobile robot. The sonar
system will ping immediately after the system is turn on. All
of the four sonar sensor will ping at the same time and sound
pulse will be sent to the environment and bounce back to the
sonar sensor. When the sonar simultaneously trigger it may
happen when the second sonar sensor receive sound pulse
information from the first sonar sensor. The data are invalid
180
2009 5th International Colloquium on Signal Processing & Its Applications (CSPA)

because of the crosstalk of the system. To overcome the Time Delay versus Time
problem is by triggering the sonar sensor one by one within
the required time but this system only applicable for one or
two sonar sensor. The more sonar the more complicated the
timing to trigger the sonar sensor.

Fig. 7. Commanded loop Daisy chaining schematic

Black: Sonar 1(s1)
Red: Sonar 2(s2)
Blue: Sonar 3(s3)
The best way to deal with this complexity is the daisy Yellow: Sonar 4(s4)
chasing application method that will concentrate on the Range: 18” where s1=1291 s2=1288 s3=1287 s4=1310
electronic circuit rather than software method. The sensors
will be chained together as shown in Figure 7 by linking the Fig. 8. Data in pulse from Sonar sensor for range 18”
TX of unit 1 to RX unit 2 an so on and the entire BW pin is
tied to high. To run the system in sequence, strobe the first Inches versus Time
sensor within 20 us and both input data can be read in
sequence. From LV-MAX-sonar sensors two outputs can be
manipulate which are AN (analog output) and PW (pulse
width output). This system will use PW as an output.
Figure 8 shows the data from 4 sonar sensors which has
been placed in line between each with 3.5 cm between each
sensor. The data show the range of 18 inches the value
approximately within range 1287 s to 1310s. The conversion
of pulse value to inches using equation (2), give data as in
Figure9 which is 17 inches. Figure 10 show data from the
sonar sensor without daisy chaining application method gave
the unpredicted value vary between 1500 s to 2000 pulse for
18 inches obstacle. The false data happen when the crosstalk
between the sonar sensors as discuss above.

Black: Sonar 1(s1)

Red: Sonar 2(s2)
Blue: Sonar 3(s3)
Yellow: Sonar 4(s4)
Range: 18” where s1=17 s2=17 s3=17 s4=17
Fig. 9. Data in inches from Sonar sensor for range 18”

978-1-4244-4152-5/09/$25.00 ©2009 IEEE 181

2009 5th International Colloquium on Signal Processing & Its Applications (CSPA)
Time Delay versus Time
Black: Sonar 1(s1)
Red: Sonar 2(s2)
Blue: Sonar 3(s3)
Yellow: Sonar 4(s4)
Fig. 10. Data in pulse from Sonar sensor without Daisy Chaining application
VI. DECISION MAKING
The microcontroller is responsible to collect all the data
from the GPS module and also from the sonar sensor. The
decision making is based on the data collection to control the
DC motor. Figure 11 show the flow chart of the system. All
the waypoints need to be preset to the GPS module before
mobile robot can be navigated. Base on NMEA code the data
are parsed and the suitable data are waypoint and heading
information. From this data the decision making can be made
whether mobile robot to move left or right to correct the error
of the system. During the navigation mobile robot also need to
avoid obstacle. If there any obstacles the mobile robot avoid
the obstacle and return to the preset path.
VII. RESULT
The mobile robot is able to move according to preset
waypoint with obstacle avoidance implemented in the system.
The accuracy of the navigation highly depend on the signal
from satellite with vary from 4 to 12 input satellites. The
accuracy of hitting the mark can drift from 10 feet to 20 feet.
If the waypoints are preset from one point to another between
12 feet and the accuracy is 20 feet the waypoint 1 and 2 will
be recognize as one point. To make system more effective due
to the less accuracy of the system the preset waypoint need to
be far from each more than 20 feet.
The obstacle avoidance can detect rock and small tree on it
path. The main issues of the obstacle avoidance system are the
detection of the grass as an obstacle can make the system
unreliable. The position of the sonar sensor needs to move
higher than the grass.
978-1-4244-4152-5/09/$25.00 ©2009 IEEE
Fig. 11. Flow chart for GPS navigation and obstacle avoidance.
VIII. CONCLUSION
This project shows that the mobile robot navigation using
GPS module can be built using combination of RC truck
between GPS and sonar sensor. The concept of the system can
be applied to bigger scale mobile robot with better accuracy of
GPS system and better obstacle avoidance system such as
vision and laser range finder.
ACKNOWLEDGMENT
The authors thankfully acknowledge the funding [Code No:
9003-00132] received from FRGS of Malaysian Government.
REFERENCES
[1] Richard W. Wall “Creating a low-cost autonomous vehicle”, IECON
02 [Industrial Electronics Society, IEEE 2002 28th Annual Conference
of the IEEE], 5-8 Nov. 2002 pp 3112- 3116.
[2] Khatib O. “Real-time obstacle avoidance for manipulators and mobile
robots”, The International Journal of Robotics Research 5(1) 1986 pp
90-98.
[3] Lino Marques. “Mobile pneumatic robot for demining”, Proceedings
of the 2002 IEEE International Conference on Robotics & Automation
Washington, May 2002 pp 3508-3513.
[4] Lewinger, W.A. “Obstacle Avoidance Behaviour for a Biologically-
inspired Robot Using Binaural Ultrasonic Sensors” Proceeding of the
2006 IEEE International Conference on Intelligent Robots and
Systems, 2007. IROS 2007, October 2006 pp 6.
182
2009 5th International Colloquium on Signal Processing & Its Applications (CSPA)

[5] Johan Borenstein. “Mobile Robot Navigation in Narrow Aisles with

Ultrasonic Sensors”, Proceeding on ANS 6th Topical Meeting on
Robotics and Remote Systems, February 1995.
[6] Mohamad Hanif Abd Hamid ”Autonomous Mobile Robot (AMRSBot)
with GPS navigation and Obstacle Avoidance System” Proceeding on
BICET 2008.

978-1-4244-4152-5/09/$25.00 ©2009 IEEE 183

View publication stats