1. 1
Technical Report, Final Year Project II, 2013/2014,
Maritime Technology Programme, School of Ocean Engineering,
Universiti Malaysia Terengganu.
A Waypoint-based Rover for Coastal Surveillance
M. A. Mohamad and A. F. Mohamad Ayob
Universiti Malaysia Terengganu, Malaysia
A coastal surveillance activity is one of the main roles of salvation which should be concern
among the worldwide nations today due to its significance to minimise unwanted incidence
along the coastline. Over the past decades, the contagion of threats such as terrorism, piracy
and criminal activities has increases by years due to adoption of infirmity salvation system
across the maritime border. Moreover, the existing of surveillance aid tools such as radar,
sensors, static location cameras and satellites is not enough to ensure the salvation of the
coastline. In this research, a thesis on the design and prototype building of a rover that able to
move by following a set of instructions by passing through waypoints is presented. The
objectives of this study is to set up a waypoint-based system that is suitable for coastal
surveillance, to build a prototype of rover that can move on any surfaces of roads and to
validate the operational aspect of the built prototype. This prototype of rover will be
equipped with the Global Positioning System (GPS), Magnetometer and Wireless
Communication in order to set up a waypoint-based system to be executed via Arduino
Programming software. Therefore, this waypoint-based rover should offer a contribution to
improve the existing monitoring aid tools system for coastal surveillance.
1. INTRODUCTION
Nowadays, the globalization linkage has not
only led to a significant increase in international trade but
also to a significant increase in cross-border crime.
International terrorism, trafficking and drug smuggling
are issues relevant to government authorities.
Furthermore, the piracy activity is reemerged as a serious
threat impeding to global business [1]. These activities
generally take place on the waterways. As a result, the
authorities need to know the identity and intentions of
unknown objects in their ports, coasts, and waterways.
As time changes, a lot of technologies emerge
which enable the improvement of people’s lifestyle such
as the usage of portable gadget and other accessories for
interact in long range. While the security aid tools such
as radar, sensors, and static location cameras particularly
used by the government authorities and non-government
agency as the adoption systems for salvation, protection
and so on. Unfortunately, some of them are misuse of
this advancement of technologies for their own sake to
conduct a criminal. This criminal activity includes
invasion of an organization and breaks security system
for malicious reasons. Moreover, due to failure to control
the maritime environment will result in unregulated area
which can be exploited for terrorist activity [2].
Instead of using a security aid tools such as
radar, sensors, and static location cameras, there are
surveillance system that recently used by authorities
which is the usage of anomaly detection. It is one of the
technique for improving the safety and security of
maritime domain. However, there is a lack of
corresponding data in the coast guard systems among the
evaluation of anomalies. The infirmity of the system is
due to inconsistent time formats in different data sources
and various settings in the transmitters [3]. Whereas, the
surveillance activity in wide area of sea poses particular
challenges using the spaceborne sensors due its infirmity
to detect small object such as boats [4]. The images
produced by spaceborne sensors shows that small
unknown objects and the surrounding seems stationary.
Moreover, the total area of the land and object on earth
cannot be accurately specified due to its resolution is
relatively high.
Generally, threat may come from sea as well as
by land, and it must be detected while far away in order
to have time to deploy appropriate forces and avoid
tragedies. In the present, the existing of surveillance aid
tools is not enough to ensure the salvation of coastline is
protected. Hence, this research project will focuses on
development of a rover prototype and might able to
increase the effectiveness of monitoring aid tools in line
with the current advancements of technologies by
designing and building a prototype of rover that will be
able to operate autonomously by following a sets of
instructions to passing through a waypoints via the
wireless communications through the computer.

2. 2
2. METHODOLOGY
This research project involved with design and
builds a waypoint-based system in the rover prototype
which is programmed via the Arduino programming for
the enhancement of coastal surveillance activity. This
study consists of a few procedures in order to achieve the
goals. The chosen study site is in Maritime Technology
Laboratory and sport complex terrain of UMT for the lab
works and testing the operations of the build prototype.
The built prototype shall operate within 10 to 20 metre of
coverage and in a wide area due to the usage of GPS,
magnetometer and wireless communication to operate
over long distances by passing through the specified
coordinates for collecting data.
2.1 Designing Tools
As a preliminary way to conduct this project,
the first steps are to design the initial sketch for the
prototype based on previous idea among researchers. The
tools involves in designing is Google SketchUp (Fig. 1)
software. The software was famously used among
architectural, engineers as well as video game designer.
This software was chosen since it is suitable to represent
or show the preliminary ideas for presentations of the
prototype. Moreover, it is easy to use since it provides a
downloadable model for modification of the built
prototype.
Fig. 1: Google SketchUp Fig. 2: Solidworks
The following tools involve in designing the
prototype is Solidworks (Fig. 2) software. This CAD
software is more complex to be used due to more
features that can be customized by the user. Moreover, it
able to create a 3D model with high detailed of
specifications such as dimension of the scale,
measurements, and precision. This software has to be
accurate enough to be used in production, construction,
or even in physical simulations. In addition, it has entire
toolsets devoted to the more artistic side of modelling
and animation, from shape to texture within the toolsets
devoted to creating seamless timeline-based animations
involving multiple objects interacting with their
environments.
2.2 Designing a Prototype
2.2.1 Conceptual Design
In order to establish a new design of rover, a
tool incorporated started with the Google sketchUp. The
initial prototype design (Fig. 3) is used to show that the
conceptual operations of the prototype by testing the
movements of the rover using a four DC motors
connected perpendicular on each vertices.
Top view Isometric view
Front view Side view
Fig. 3: Conceptual Design
2.2.2 Embodiment Design
Once the idea of prototype was design, the further
development on the rover can be continued. In this
project, the used embodiment prototype is from the
modifications of the available rover. However, the
available prototype needs to be modified according to the
criteria from previous conceptual design which uses four
DC motors. Fig. 4 below shows that the needed parts to
be modified for the housing of DC motors and Fig. 5 is a
prototype of the available rover.
Front base
Rear base
Fig.4: DC motor base to be modified
Fig. 5: Prototype of available rover

3. 3
2.3 Incorporated Tools
2.3.1 Arduino UNO Microcontroller
The main electronic components used to execute
the instructions of a waypoint-based system are the
Arduino UNO microcontroller as shown in Fig. 6 below.
It is a microcontroller board which is an open-source
physical computing platform based on a simple I/O board
and a development environment that implement the
processing or wiring language. Arduino can be used to
develop an object such as motors, servos, sensors and
another varies of electronics components to operates on
specified tasks by executes a set of instructions declared
via the byte codes.
Fig. 6: Arduino UNO Microcontroller
2.3.2 DFRobot Motor Shield
Another types of microcontroller used is
DFRobot Arduino motor shield as shown in Fig. 7 below.
This motor shield allows Arduino to drive two channel
DC motors. It uses a L298N chip which delivers output
current up to 2A each channel. The speed control can be
achieved through conventional of which can be obtained
from PWM output pin.
Fig. 7: DFRobot Motor Shield
2.3.3 GPS Module
In order to locate a waypoint, the types of used
GPS module are Skylab SKM53 Series (Fig. 8) with
embedded GPS antenna which enables high performance
navigation in the ambiguous visibility of environments.
The 6-pin and USB connector design is the easiest and
convenient solution to be embedded in a portable device
and receiver car holder, personal locator, speed camera
detector and vehicle locator.
Fig. 8: Skylab SKM53 GPS
2.3.4 Magnetometer (Compass)
Additional components involved in set up a
waypoint-based system are the adoption of compass or
magnetometer as shown in Fig. 9 below. The type of
magnetometer used is Triple Axis Magnetometer
HMC5883L. It is used to calculate the angle to the
desired waypoints by referring the heading or Y-axis.
Fig. 9: Triple Axis Magnetometer HMC5883L
2.3.5 APC220 Radio Data Module
The communications tool used in the rover
prototype is a APC220 Radio Data Module (Fig. 10). The
module provides a simple and economic solution to
wireless data communications. It Transmit signal
communications up to 1000 meters which is feasible to
be used for long range of interactions particularly for the
waypoint-based systems.
Fig. 10: APC220 Radio Data Module
2.3.6 DC Geared Motor
The type of DC motor used is DFRobot Micro
DC Geared Motor with Back Shaft (Fig. 11) which is
ideal for DIY enthusiasts and is designed to easily
incorporate the DFRobot Encoder. These motors are easy
to install, small, and ideally suited for use in a mobile
robot car.
Fig. 11: DC Geared Motor

4. 4
2.4 Mathematical Concept
In order to create a waypoint-based system, the
related mathematical concept used is comes from the
Pythagoras theorem which is the appropriate way to
calculate the distance between two waypoints. Fig. 12
below illustrates that how this theorem is applied to
enable the rover autonomously move to the desired
points. Initially, before the rover start moving to the
target waypoint, it will calculate the angle to desired
position based on the heading which facing the North
Pole.
N wpt2
𝑥
wpt1
N 𝑦 wpt2
𝑥
𝑙
wpt1
𝑙
𝑦
𝑥
𝑙2
= 𝑥2
+ 𝑦2
𝑙 = 𝑥2 + 𝑦2
tan 𝜃 = 𝑦/𝑥
Fig. 12: Conceptual application from Pythagoras theorem
3. RESULT AND DISCUSSION
In this research, all the results consist of
locating waypoints for path planning of the rover via the
GPS. After the waypoint was located, the further step is
to execute the created source codes in the Arduino
programming for operations of autonomous system. All
the information will be discussed based on the research
objectives which are to create an autonomous system for
the application of coastal surveillance activity. Secondly
is designing and building a prototype of rover based on
the conceptual design. Lastly is the testing of operations
of the built prototype.
3.1 Design and Build of a Prototype
Based on the previous conceptual prototype
design, the rover should be installed with four DC geared
motor. However, In order to adopt the conceptual design,
the available rover was modified by replacing the front
and rear base for the housing of DC geared motors. Fig.
13 below shows that the new front and rear base design
via the use of Solidworks software was made from the
product of 3D printer. Hence, the conceptual prototype
can be implemented on the embodiment prototype as
shown in the Fig. 14 below.
Front base
Rear base
Fig. 13: Product of 3D Printer
Fig. 14: Conceptual and Embodiment Prototype
3.2 Locating the waypoints
The coordinate for the waypoints is located
using a GPS for the path of the rover to move
autonomously. In this project, the study site for testing
the operations of the prototype is in Sport Complex
Terrain of UMT. This place was chosen because of its
suitability of conditions due to wide area and quite far
from crowdedness of civilian. Moreover, this area could
facilitate the activity for testing the conceptual operations
of the waypoint-based system before it will be fully
adopted in the coastline area. Fig. 15 below illustrate that
the image from satellite for the located waypoints via the
Google Earth software. The distance point was
approximately set up from 10 to 20 meters between each
point.
Fig. 15: Located waypoints from satellite image

5. 5
3.3 Algorithm of the Waypoints-based System
The following Fig. 16 below is a flowchart
process of the algorithm waypoints-based system for the
rover to be executes on the applications of coastal
surveillance activities.
Starting from initial waypoint
Calculates heading of current location to the target waypoints
Rotates the rover to desired angle
Measure the distance from rover to the waypoint
Move towards the waypoint for 5 seconds
Stop as well as reach the destination
Fig. 16: Algorithm of the Waypoints-based System
3.4 A Waypoint-based System code
The combination of GPS, magnetometer, motor
shield, Arduino microcontroller and wireless
communications module is used to execute the waypoint-
based system code. The connections for magnetometer
was set up of which is Arduino (GND) to HMC5883L
(GND), Arduino (3.3V) to HMC5883L (VCC), Arduino
(A4) to HMC5883L (SDA) and Arduino (A5) to
HMC5883L (SCL). While the wireless module and GPS
shares the same pin in order to minimise the usage of
bunch of jumper wire and to make the circuit looks tidy
of which is Arduino (TXD) to GPS (RXD), Arduino
(RXD) to GPS (TXD), Arduino (5V) to GPS (VCC) and
Arduino (GND) to GPS (GND). For the operations of
DC motors, the pin 4, 5, 6, and 7 is used to regulate the
Pulse Width Modulation (PWM) and speed control
which can be adjusted; 0 for stop and 255 for maximum
speed.
3.5 Analysis of Global Positioning Systems (GPS) errors
While testing the operation of the waypoint-based system, there are some error occurs on the value of the distance
between the point of which the rover stopped and the target waypoint due to GPS orbital errors, satellite clock errors,
and ionosphere effects change over time which can lead to variations in the stability of the estimated GPS positions [5].
Table 1 below shows that the number of experiment of a waypoint-based system is repeated for ten times in order to
obtain the percentage errors of the GPS.
No.
of
Exp.
Initial Waypoints
(WPT1)
Target Waypoints
(WPT2)
The points of which the rover
stopped 𝑳 𝒔
(m)
𝑳𝒊
(m)
𝑳 𝒔𝒕
(m)
Latitudes Longitudes Latitudes Longitudes Latitudes Longitudes
1 5.4053502 103.0902175 5.4055099 103.0901870 5.40547990 103.0901565 15.92 18.0 2.08
2 5.4053502 103.0902175 5.4055099 103.0901870 5.4054899 103.0902099 15.56 18.0 2.44
3 5.4053502 103.0902175 5.4055099 103.0901870 5.4054698 103.0901107 17.79 18.0 0.21
4 5.4053502 103.0902175 5.4055099 103.0901870 5.4054698 103.0901489 15.31 18.0 2.69
5 5.4053502 103.0902175 5.4055099 103.0901870 5.4054698 103.0901423 15.69 18.0 2.31
6 5.4053502 103.0902175 5.4055099 103.0901870 5.4054399 103.0901336 13.63 18.0 4.37
7 5.4053502 103.0902175 5.4055099 103.0901870 5.4054601 103.0901589 13.84 18.0 4.16
8 5.4053502 103.0902175 5.4055099 103.0901870 5.4052201 103.0902099 14.49 18.0 3.51
9 5.4053502 103.0902175 5.4055099 103.0901870 5.4054101 103.0903328 14.40 18.0 3.60
10 5.4053502 103.0902175 5.4055099 103.0901870 5.4053101 103.0903175 11.93 18.0 6.07
Table 1: Experiment of a Waypoint-based System
Where; Ls is the distances between initial points to the point of the rover stopped, Li is the initial distances from current
waypoint to the target waypoint, and Lst is the distances between the points of the rover stopped to the target waypoint.

6. 6
From Table 1, the percentage errors of GPS for each
number of experiments can be calculated using the
following equations.
𝐸𝑟𝑟𝑜𝑟 𝑜𝑓 𝐺𝑃𝑆 =
𝐿 𝑠𝑡
𝐿𝑖
× 100 %
Based on the calculations of the percentage errors of GPS
for each experiment, all of calculated data is tabulated as
shown in the Table 2 below.
No. of Experiment Percentage errors of GPS
1 11.56
2 13.56
3 1.17
4 14.94
5 12.83
6 24.30
7 23.11
8 19.50
9 20.00
10 33.72
Table 2: Percentage errors of GPS
The mean, 𝑋 for the percentage of GPS error and the
mean, 𝑋 for distances between the points of the rover
stopped to the target waypoint can be calculated as
shown in the formula below.
𝑋 =
𝑋
𝑛
Where, 𝑋 is sample number, 𝑛 is number of samples.
Both of mean value is ± 17.50 % and ± 3.14 𝑚
respectively.
While the standard deviation, σ for the percentages of
GPS error and the distances between the points of the
rover stopped to the target waypoint can be calculated
using the following equations.
𝜎 =
(𝑋 − 𝑋)2
𝑛 − 1
Where, 𝑋 is sample number, 𝑋 is sample mean, 𝑛 is
number of samples. Both of mean value is 8.80 % and
1.58 𝑚 respectively.
Fig. 17 below shows that the line graph of the percentage
errors of GPS according to the tabulated data from Table
2 above during validation of the waypoint-based system
in ten times of repetition. The obtained result shows a
fluctuation values from the beginning of experiment until
the last one.
Fig. 17: Line graph of the percentage errors of GPS
There are number of factors that influence the accuracy
of GPS position including the satellite, the receiver, and
signal propagation errors. The total common positional
accuracy for a GPS receiver without correction is in the
order of 10 meters depending on its operating condition
[6]. Besides that, the sources that impact positional
accuracy including ephemeris error, satellite and receiver
clock error, multipath error, receiver measurement noise,
satellite geometry measures, tropospheric delay, and
most significantly is ionospheric delay.
4. CONCLUSION
This research project could contribute to the
enhancement of the efficiencies of maritime border
salvation since the usage of road vehicles is highly
effective for monitoring the environment nearby due to
close with surrounding conditions. In other words, the
effectiveness of using land vehicles might be suitable
since the acts to employ a force within the incoming
threats will take an action immediately
5. REFERENCES
[1] Bakir, N. O. (2007). A brief analysis of threats and
vulnerabilities in the maritime domain.17, pp 198-207.
[2] Austin, S. J. (2012). Regeneration of United Kingdom
wide-area maritime patrol capability. 3, pp 221-228.
[3] Kazemi S, Abghari S, Lavesson N, Johnson H, Ryman P
(2013). Open data for anomaly detection in maritime
surveillance. 40, pp 5719-5729.
[4] Greidanus, H. (2008) Satellite imaging for maritime
surveillance of the European Seas. 7, pp 343-358.
[5] Olynik M, Petovello M G, Cannon M E, Lachapelle G
(2002). Temporal impact of selected GPS errors on point
positioning. 3, pp 47-57.
[6] Klobuchar, J.A. (1996). Ionospheric effects on GPS: In
Global Positioning System. 2, pp 485–516.
0
10
20
30
40
1 2 3 4 5 6 7 8 9 10
Percentageerrorsof
GPS
No. of Experiment