Abstract— This paper will discuss about Intelligence Pothole Detection Sensor
and Monitoring System using smartphone technology which can alert all the
driver in avoiding potholes on the road by giving them warning through
smartphone device and car sensor. If the car approaching the pothole, the
sensor will detect it and give warning alarm to driver so that driver can avoid
it. Data from car sensor will then relate and transfer the information with an
application in driver phone to restore the data about the pothole. Driver also
can report the pothole to the right authorities using data collection using
that application. It can be used in government sector for them to know about
the pothole that occur in certain area.
car sensor; style; pothole detection system; government
With the increasing number of vehicle in
the road, there has been increasing load on infrastructure such as road and the
traffic is growing day by day too. In cities across the world, ministries,
department and agencies are concerned with road maintenance. They had to spend
huge amount of money yearly in maintaining and repairing the road such as
potholes. Worsened road condition are one of the increasing problem that road
has facing. This is because of disruption in the
surface of a roadway where a portion of the road material has broken away,
leaving a hole 1. Rains, oil spill and industry lorry which is big are
one of the reason causing the pothole and because of this road condition, it
can cause an accident too. These become very important to get the information
for the bad road condition which is collect the information, reporting to the
authorities, distribute to other drivers so that it can be warn and monitoring
the pothole using the system. It is important to make the system is attractive
and user friendly so that system can be accepted by wide user community.
II. problem background
The system consist two component which is
one is mobile device and other is the sensor on the car. Mobile device using
the application will store the data on the database. Government had always have
an issue with this pothole. Sometimes, they didn’t know that the pothole has
been there for a long time. With this system, user can report the pothole that
has been detect by them to the government and authorities. At the same time,
when car sensor detecting the pothole, user or driver can avoid it and it might
save them from any incident. This device or sensor are responsible to for
warning the driver about occurrence of the pothole.
III. research background
The current and past trend on this
research area are the evolution from installing the device or sensor in
vehicles to using the sensor and data in smartphone. Post processing to real
time detection, machine learning approaches and threshold based detection have
been using for this research about pothole detection system.2. The rise of
mobile and sensor technology has been inspired by all this and this research is
to aim that everyone can use it and introduces the improved techniques. Based
on Mohan, P. et.al in 3, he was the first to document a system using
smartphone sensor. Together with data from the accelerometer alone, it use to
implement a virtual re orientation and in addition made use of the magnetic
vector values obtained from the magnetometer sensor. In 4, it stated that
pioneer in real-time detection of road conditions and this was made possible
and effective with their classification algorithms: Z-Diff and Z-Tresh. Accelerometer
and GPS for data collection are mostly use for this method. Some of this method
too are using machine-learning algorithm.
IV. Literature review
A. Method using Specialize Sensor
There are several research and method
that are using specialize sensor that put in the vehicle. Below table are
Research Model / Propose
Distributed mobile sensor
computing system called Cartel 5.
Collecting and process data will be send to portal
based upon the continuous queries which are processed by continuous query
processor on remote nodes and it’s include
a set of sensors installed in vehicles.
like GPS for monitoring the movements of vehicles.
CarTel includes, CafNet a networking stack that uses
opportunistic connection (e.g. Wi-Fi, Bluetooth) to transfer information
between portal and remote nodes.
currently does not offer a way to aggregate
information gathered across different users and it does not include machine
learning; it just replies to the queries based upon the data stored in
3-axis accelerometer and GPS mounted on the dashboard
to monitor road surface
Also differentiate potholes from other road anomalies.
collects the signals using accelerometer and use
machine-learning algorithms to identify potholes
signals are then passed through a series of signal
processing filters, where each filter is designed in such a way that it will
reject one or more non-pothole events (manholes, expansion joints, railroad crossing)
Collects the sensor data using three-axis
accelerometer and GPS
Sensor data has 4-tuples: current time, location,
velocity and three direction accelerations.
Cleaning the data before processing or analyzing it to
deal with technical challenges like GPS error, and transmission error
Analyses the Power Spectral Density (PSD) to detect
pavement roughness using Fourier transform.
The International Roughness Index (IRI) is calculated
based upon PSD. The pavement roughness is then classified in four levels
(excellent, good, qualified and unqualified) according to, the Technical Code
of Maintenance for Urban Road CJJ36-2006, one of the industry standards in
the People’s Republic of China.
The system provides the evaluation of a section of
road based upon its roughness. However, this system does not provide the
proper location of pothole, bump or manhole.
B. Method Using Smartphone Sensor
There are some
research using the method of smartphone sensor too. Below table are refer.
Model / Propose System
Rich monitoring of
road and traffic conditions using mobile smartphones 2
Detect potholes, braking, bumps and honks using
accelerometer, microphone, GSM radio and GPS sensors present in
Triggered sensing where a high energy-consuming sensor e.g.
GPS, microphone, is activated by a low energy-consuming sensor e.g.
accelerometer, or cellular radio making the system energy efficient.
Relevant location can be tagged with sensed information such
as honking or bump, and the researchers employed GSM radios for
energy-efficient localization by using strongest signal (SS)-based
This system uses an algorithm based upon Euler angles for
reorientation. The sensor is virtually rotated along the vehicle’s axis
using pre-rotation, tilt and post rotation angles (Euler angles).
Real time pothole detection using android
smartphone with accelerometers 4
Detects events in real-time and collects the data for
Data is collected using 3-axis accelerometer sensor present
Four algorithms to detect a pothole. The first two algorithms
(ZTHRESH and Z-DIFF) are for real-time detection and the other two
(STDEV (Z) and GZERO) are used for off-line post-processing of data.
This system gives a true positive result of 90% (approx.).
Uses smartphone sensors for traffic state monitoring and
detection of bumps.
It uses accelerometer sensor to collect the data. The device
(phone) is to be reoriented as it can have any arbitrary orientation
when kept inside the vehicle.
This system reorients the phone in two steps using
accelerometer and magnetometer.
Phone’s axes are aligned with geometric axes. A rotation
matrix is formed using Gravity Vector given by accelerometer and
Magnetic Vector given by magnetometer. This rotation matrix represents
the angles of rotation of device’s axes to align with geometric axes.
This system detects two events i.e. braking and bump. The
bump event is detected by the standard deviation on window of one-second
duration with sampling rate of 50 readings per second over the z-axis
value. The braking event is detected by using the difference between the
maximum and minimum value within a window for y-axis value.
This system gives 10% false negative rate for bump detection
and 21.6% false negative rate and 2.7% false positive rate for braking
Mobile phone sensor to detect driving
Mobile phone application that uses GPS, accelerometer and
microphone to collect the data.
Detects road and traffic conditions along with driving
This application is used to detect various events based upon
the patterns observed and does not use machine learning.
Completely based upon the patterns obtained from the sensor
Intelligent Smart Selangor by
collaboration between Selangor local authorities and Google Asia Pacific under
the Waze Connected Citizen Programme, since 2016, Selangor Motorist can file a
complaint about pothole via Waze. Menteri Besar Selangor Incorporated has
develop Intelligent Smart Selangor 12 local council in Selangor. It tracks the
workflow of local councils, down to the officer in charge. Dr Fahmi Ngah, from
Smart Selangor Delivery Unit said that, collaboration with Waze as good because
it leveraged on reports submitted by Waze user.
Through Selangor’s Intelligent Response
system, the Mayor and the President of the Municipal and District Council no
longer need to roam along the road to monitor if there is a pothole on the road,
because user can directly report it through the Waze and officer in charge will
get a notification about the report. With this system, a problem regarding
pothole can be solved more effectively and quickly. It is user friendly so that
user able to use it without any problem since all user are using the Waze and
they are familiar with the application.
According to Dr Fahmi, almost a year
later since its inception, the response has been really great and repair
efficiencies have improved more since the system start to use by user. When
user saw a pothole on the road, user will directly report the pothole using
their Waze. The report will go through into this Intelligent System Selangor,
and officer will get notified using this system or application that they
install in their mobile device. They figured that from Feb 2017 to July 2017,
5173 valid reports were received. About half were resolved within 5 days. Their
challenge now are to extract that data from Waze in an automated and regular
manner, process it and pass it to the local council and subsequently, the
person who is in charge of the patching job11.
Intelligent Response Selangor Interface
suggestion on future directions
experience with variety of the scenario will be propose in the future. On the
next step, sensor will be attach on a real vehicles and measure their response.
Different scenario like pothole on the slopes, turns will be apply and to see
how the sensor readings such condition. Simulation for this will be apply too
to check if this is okay to use for the future. Based on the implementation of
Intelligence Pothole Detection Sensor and Monitoring System using Smartphone
Technology in Malaysia, the following recommendation are made:
More representative and larger data need to be collected
Traffic exposure data had to be obtain accurately as the
current challenge of road traffic anomalies accident.
More efficient tool for representing traffic flow and
visualize traffic on specific routes that have a pothole so it can be avoid
The future use on the intelligence pothole
detection sensor and monitoring system using Smartphone technology in Malaysia
is the system can enable the car sensor to detect, predict and share data. This
the intelligence pothole detection sensor and monitoring system using
Smartphone technology will make user vehicle to collect the data about the
pothole location and status of potholes, manhole covers and broken drains and
will also enable the vehicles to send and receive the warning then will make a
driver slow down or the car adjust its suspension setting to smooth the ride
and reduce the impact. It will then can reduce the potential for punctures and
vehicle damage. When a driver avoided the pothole, it will make the care more
safety. From the sensor, it will connect to user smartphone to report on the
pothole and share date to others driver too so that others know about the
pothole. Authorities then will get a report from the request and will do the necessary.
This paper studied on an application of
the mobile for pothole detecting system, detecting and reporting the surface
condition of the road and report it to the right authorities. This method are
using sensor that place on the car that detect the pothole and data from the
sensor integrated with mobile phone application. This can be used to help the
government on how they manage the pothole using this system. Moreover, a
pothole detection approach are to make them easier to manage and immediately
repair the pothole using this apps. It can reduce the accident that occur
because of the pothole.