2017 International Conference on Innovations in information Embedded and Communication Systems (ICIIECS)

Road Quality Management System using

Mobile Sensors

Arun Kumar G 1 Santhosh Kumar A 2

UG, Computer Science and Engineering UG, Computer Science and Engineering
Kamaraj College of Engineering and Technology Kamaraj College of Engineering and Technology
Virudhunagar, India. Virudhunagar, India

Ajith Kumar A 3 Maharajothi T 4

UG, Computer Science and Engineering Assistant Professor
Kamaraj College of Engineering and Technology Kamaraj College of Engineering and Technology
Virudhunagar, India. Virudhunagar, India

Abstract—One of the major problems in developing countries safely. With the proposed system an attempt has been made to
is maintenance of roads. Well maintained roads contribute a endorse drivers to ward off the accidents caused due to
major portion to the country’s economy. Identification of road potholes and raised bumps.
distress such as potholes and bumps, helps drivers to avoid
accidents, vehicle damages and it also helps authorities to II.PROBLEM STATEMENT
maintain roads. In this paper, we discuss the previous pothole
detection methods that has been developed and proposes a cost- Road is a vital part of people’s day-to-day lives. When road
effective solution to identify the potholes and bumps on roads. In is put into use after construction, it will lead to develop various
our system, mobile phone sensors are used to identify the anomalies due to continuous rolling under the wheels, and
potholes and bumps. The proposed system captures the snow, rain and other natural factors. This will affect the quality
geographical location of the potholes and bumps using GPS of driving. With the availability of information regarding the
sensor in the Mobile phone. These sensed-data are send to the road conditions, road users can be cautious about or avoid the
cloud storage for further processing. This serves as a valuable bad roads. It is desirable to have a mechanism for detecting the
source of information for the government authorities and vehicle
condition of roads and get them repaired as soon as possible. .
drivers. An android and web application can be used to display
the road condition in the map. The road bumps and potholes are kind of conditions which are
needed to immediate monitoring and sharing. Bad roughness
Index Terms—Road Quality System, Pothole detection, condition may cause uncomfortable ride, extra fuel
Pavement management consumption, unexpected vehicle maintenance costs and
safety. Road infrastructure maintenance and management is
I.INTRODUCTION very important to keep the infrastructure in good condition so
that it does not affect road users especially in terms of vehicle
INDIA, the second most populous Country in the World
maintenance, fuel consumption and ride quality.
and a fast growing economy, is known to have a gigantic
network of roads. Roads are the dominant means of III.OBJECTIVE
transportation in India today. They carry almost 90 percent of
The main objective of this system is to design an
country’s passenger traffic and 65 percent of its freight.
application to know about the road quality. Nowadays we all
However, most of the roads in India are narrow and congested
use Google maps and its application for navigation during
with poor surface quality and road maintenance needs are not
travelling, but these applications couldn’t able to tell you any
satisfactorily met. No matter where you are in India, driving is
road’s condition or its complexity. Road Quality Management
a breath-holding, multi-mirror involving, potentially life
System using android phone proposes to utilize the GPS
threatening affair. Pathetic condition of roads is a boosting
system of phone and different sensors like accelerometer,
factor for traffic congestion and accidents. Roads in India
magnetometer, etc. of android phone, so we could analyze the
normally have speed breakers so that the vehicle’s speed can be
road and can upload this information of that road on central
controlled to avoid accidents. However, these speed breakers
server so every application user can use this information during
are unevenly distributed with uneven and unscientific heights.
traveling. This information can be helpful to user at the time if
To address the above mentioned problems, a cost effective
there are multiple routes and for destination and he can choose
solution is needed that collects the information about the
one of the finest and shortest route
severity of potholes and humps and also helps drivers to drive

‹,(((
2017 International Conference on Innovations in information Embedded and Communication Systems (ICIIECS)

IV.CHALLENGES monitoring can also help to predict the estimated arrival time
Although a lot of effort has been done to detect road from one place to another. This paper presents a detailed
conditions, a lot of methods/systems have been developed survey of methods for detecting road conditions. Form the
using Smartphone sensors, a highly reliable method is yet to be survey, it is noted that the most commonly used sensors
built. Real time road anomalies detection is so challenging that accelerometer and GPS. Smartphone sensors are gaining
none of the methods using Smartphone sensors can address it importance in this field as they are cost effective and also
alone completely. Some of the challenges are increase scalability. Analyzing form the research activities, it is
certain that this area will gain more importance in recent
A. Vibration patterns of sensor data future. There are several research issues that can be explored
A given pothole or any other road anomaly may not for improvement in existing methods and develop a highly
necessarily give the same pattern during each drive over it. The reliable method
sensors readings depend upon the speed of the vehicle, how it
approached the road anomaly and the position of the sensor i.e. VI.LITERATURE SURVEY
its orientation. It also depends upon the suspension system of Yulu Luke Chen et al., proposed a system called
the vehicle, if the suspension system is not in normal condition, Inexpensive Multimodal Sensor Fusion System for
sensors will show more deviation based on the large vibration Autonomous Data Acquisition of Road Surface Conditions [1]
experienced by the vehicle. which is a field application of a novel, relatively inexpensive,
vision-based sensor system employing commercially available
B. Benign events off-the-shelf devices, for enabling the autonomous data
There are many events that are not considered as road acquisition of road surface conditions. Detailed evaluations and
anomalies such as expansion joints, railroad crossings, door enhancements of a variety of technical approaches and
slams etc. These events are to be differentiated from the algorithms for overcoming vision-based measurement
potholes. An efficient system is to be developed that is able to distortions induced by the motion of the monitoring platform
classify different events effectively. were conducted. It is shown that the proposed multi-sensor
system, by capitalizing on powerful data-fusion approaches of
C. GPS error
the type developed in this paper, can provide a robust cost-
GPS provides the longitude and longitude values of a effective road surface monitoring system with sufficient
location. It is used to detect the location of potholes to users. It accuracy to satisfy typical maintenance needs, in regard to the
has an error of 3.3 meters. This error must be minimized to get detection, localization, and quantification of potholes and
the exact location where events to be detected have occurred. It similar qualitative deterioration features where the
may also be possible to miss some GPS data in urban areas measurements are acquired via a vehicle moving at normal
among tunnels and tall buildings. A method to minimize the speeds on typical city streets.
localization error is still an open problem. Rajeshwari Madli et al., proposed a system called
D. Network overload and delay Automatic Detection and Notification of Potholes and Humps
on Roads to Aid Drivers [2] which is used to detect the
The sensor data is to be uploaded on the server at back-end.
pavement distress such as potholes and humps not only helps
If a large amount of data is needed to be send over the network
drivers to avoid accidents or vehicle damages, but also helps
it may lead to network congestion leading to delay or loss of
authorities to maintain roads. This paper discusses previous
data. Therefore, the application is needed that must keep the
pothole detection methods that have been developed and
network usage to minimal. It will also save the communication
proposes a cost-effective solution to identify the potholes and
cost.
humps on roads and provide timely alerts to drivers to avoid
E. Privacy accidents or vehicle damages. Ultrasonic sensors are used to
The application requires the location of the device to detect identify the potholes and humps and also to measure their
the location of road anomaly and can be retrieved by user; depth and height, respectively. The proposed system captures
hence it may lead to privacy breach. There is a need of an the geographical location coordinates of the potholes and
application that is able to hide user identification to maintain humps using a global positioning system receiver. The sensed-
the privacy of user. data includes pothole depth, height of hump, and geographic
location, which is stored in the database (cloud)
V.SOCIAL IMPACT Juan Martin Raya Bahena et al., proposed a system called
India is the second largest country with huge number of Speed booms detection for a ground vehicle with computer
population and vehicles. In India most of the communications vision [3] which is a new stereo vision system for ground
are being dependent on road transport only. Often the road vehicle to detect speed bumps. The damage for cross speed
accidents may occur due to Reckless Driving and improper bumps fast can result in harms to people breakdowns to
road conditions. Government of India has taken necessary steps vehicles. In this paper we propose speed booms detection
to maintain the road conditions. Even though the technology method using Disparity, Border detection, Morphological
has improved a lot there will not be of accuracy. Due to heavy Image processing, canny edge detector. The system was built
rain the road conditions are being damaged easily. Road using two web cameras. The importance of this work is to
2017 International Conference on Innovations in information Embedded and Communication Systems (ICIIECS)

present a methodology to detect speed booms in a ground A. Sensing

vehicle. In the broadest definition, a sensor is an object whose
Jakob Eriksson et al. proposed a system called Pothole purpose is to detect events or changes in its environment and
Patrol system [4] uses 3-axis accelerometer and GPS mounted sends the information to the computer. Today sensors are used
on the dashboard to monitor road surface. It not only identifies in everyday objects. Our mobile phone contains so many
potholes but also differentiate potholes from other road sensors. We had used the accelerometer and GPS sensors in the
anomalies. It collects the signals using accelerometer. It uses mobile phone to detect the pothole and to find out the current
machine learning algorithm to identify potholes. These signals location and speed of the vehicle. We can detect the potholes
are then passed through a series of signal processing filters, by observing the accelerometer data patterns. If we identified
where each filter is designed in such a way that it will reject the pothole, immediately we access the location of the pothole
one or more non-pothole events. by using the GPS sensor in the mobile phone. We can also find
Chen et al., proposed a system called RCM-TAGPS [5] the speed of the vehicle by using the GPS values. Fig.2 shows
which is used to collect the sensor data using three-axis the flow chart for sensing the data sets from the mobile phone.
accelerometer and GPS. The sensor data has 4-tuples: current
time, location, velocity and three direction accelerations. This
system also does the data cleaning before processing or
analyzing it to deal with technical challenges like GPS error,
transmission error.
Mednis et al., proposed a system called Real time pothole
detection using android smartphone with accelerometers[6]
which uses Android OS based Smart-phones having
accelerometer sensor for detection of potholes in real time.
This system detects events in real time and also collects the
data for off-line post-processing. The data is collected using 3-
axis accelerometer sensor present in Smart-phones.
Bhoraskar, R et al., proposed a system called Wolverine:
Traffic and Road Condition estimation using Smartphone
Sensors Wolverine [7] method 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.
VII.SYSTEM ARCHITECTURE
The basic system architecture of our project contains 4
tasks sensing, storing, analyzing and retrieving the data. In the
first task, we acquire the data from the mobile phone sensors.
The second task is to send the sensed data to the Cloud storage.
The third task is to analyze the cloud data to predict the road
conditions. The fourth task is to retrieve the data and display
the condition of the road as the markers in the map, the user
can easily be able to access the data from the various devices.
Fig. 1 shows the basic system architecture of our system.

Fig. 2. Flow chart for sensing the data

This module helps in finding bumps and potholes on the
road using accelerometer and GPS sensors. Now-a-days
everyone has a smartphone with them. We can collect the data
set from everyone using the application. We created an android
application to detect the bumps and potholes in the road by
using the in-build accelerometer and GPS sensors in the
mobile. This application detects the potholes or bumps by
observing the sudden changes in the accelerometer values. A
standard value is considered as a threshold value for detecting
the bumps and potholes. If the accelerometer difference value
Fig. 1. System Architecture
2017 International Conference on Innovations in information Embedded and Communication Systems (ICIIECS)

crosses the threshold value, immediately the location and the D. Retrieving
speed of the vehicle was recorded using the GPS sensor. Retrieving is the fourth task, here the datasets present in the
B. Storing public cloud are accessed first. The location of the potholes are
plotted as markers based on the latitude and longitude values
Storing is the second task performed after all
present in the cloud storage. This module provides the user
the data sets are sensed from the mobile devices. In this
interface for showing where the potholes are located. A web
subsystem, the user send the sensed data to the cloud storage.
application is used for showing the plotted map to the user. The
This can be said as the backbone of the system. This subsystem
user can easily view these data from any device.
collects the data from different vehicles and Co-ordinates the
This module provides the view to see the data stored in the
data. Since our domain is Internet of Things (IOT), it requires
centralized data storage. The user gets the details of the road
the uninterrupted network support for sending the data to the
from the web application. This web application will display the
cloud storage. Here all the data sets are send to the public cloud
road conditions in the map. Road conditions are marked as the
storage and it is further used for analyzing process. The sensed
different color markers on the map. If the user needs to choose
data sets from the first task are send to the cloud storage using
between the routes, he/she can check this web application from
the HTTP request using either GET or POST method.
anywhere to check the condition of the road.
C. Analyzing
VIII.METHODOLOGY
Analyzing is the third task that is done after all the datasets
are send to the cloud storage. Here the sensed data are To ensure road surface quality it should be monitored
compared to the other values that are already available in the continuously and repaired as necessary. The optimal
cloud storage. Based on the comparison the quality of the road distribution of resources for road repairs depends on the
is predicted by the web application. This subsystem is availability of comprehensive and objective real time data
responsible for determining the road conditions and adds the about the state of the roads. Participatory sensing is a
result into database. Other main function of this subsystem is to promising approach for such data collection. This system
provide data to the user as per user’s requirements. It collects presents the development of a cost-effective, sensor system
all the data send by the different mobiles. The centralized web using commercially available off the-shelf devices for
application is running on the cloud to analyze the data before it autonomous data acquisition of road surface conditions. This
gets stored in the database. There may be many chances for the system is a easy-to-use sensor system that is relatively
data received to be false values. So we have to check whether inexpensive, which could be easily mounted on vehicles, to
the values coming from the application are all valid or not. The perform periodic pavement data collection, and maintains data
challenges we referred in the chapter 1 are resolved in this in a main server. This system can be mounted on any vehicles
module. Most of the data anomalies occur because of the false using a clamp. Such frequent data gathering can provide an
vibration pattern of the mobile, benign events, GPS error, improved understanding of the formation and evolution of
network overload and delay. We have to avoid these kind of defects in time, thus leading to earlier measure to curtail the
false values by analyzing the data before storing it in the degradation of the pavement surface.
central database. Data module plays the vital role in our Our system consists of 3 subsystems: Sensing the data,
system. Fig. 3 shows the flow chart for analyzing and Data processing and plotting the data. These three subsystems
displaying the data stored in the cloud. work independent of each other, but have one center point they
revolve around; that is data. All the data sets are collected at
the initial subsystem and gets processed at the second one and
then finally gets displayed at the third subsystem.
A. Sensing the data
This subsystem is responsible for getting the data. The data
in this case would be the data about pothole e.g. location of
pothole, the severity of the pothole. Sensors are widely used in
all other fields. In our system sensors are used to predict and
test the road conditions. Using various sensors, we can predict
the location of the pothole by using the two methods: Vision
based method and Vibration based method. In Vision based
method, cameras are used as sensors to predict the road
condition. The camera captures the images in real time. These
images are applied to image processing algorithms to predict
the presence of the pothole. This process will take more time
for processing. In Vibration based method, accelerometers are
used as a sensors. We can characterize the pothole on the basis
Fig. 3. Analysing and displaying the data in the cloud of the magnitude of change in reading of accelerometer. It
consumes less time and power than the vision based method.
2017 International Conference on Innovations in information Embedded and Communication Systems (ICIIECS)

1) Accelerometer Sensor The question is about an interplay of accuracy and precision.

An Accelerometer Sensor is a device that measures Proper As a general principle, you don't need much more precision in
Acceleration. Proper acceleration is not the same as coordinate recording your measurements than there is accuracy built into
acceleration. Accelerometers have multiple applications in them. Using too much precision can mislead people into
industry and science. Highly sensitive accelerometers are believing the accuracy is greater than it really is. Generally,
components of inertial navigation systems for aircraft and when you degrade precision--that is, use fewer decimal places-
missiles. Accelerometers are used to detect and monitor -you can lose some accuracy. Accordingly, if your accuracy
vibration in rotating machinery. Accelerometers are used in needs are, say, give or take 10 meters, than 1/9 meter is
drones for flight stabilization. Coordinated accelerometers can nothing: you lose essentially no accuracy by using six decimal
be used to measure differences in proper acceleration, places. If your accuracy need is sub-centimeter, then you need
particularly gravity, over their separation in space; i.e., gradient at least seven and probably eight decimal places, but more will
of the gravitational field. This gravity is useful because do you little good.
absolute gravity is a weak effect and depends on local density Table 1. Decimal degree meaning of a GPS value
of the Earth which is quite variable. Single and multi-axis Decimal Decimal
Distance
models of accelerometer are available to detect magnitude and places degrees
direction of the proper acceleration, as a vector quantity, and 0 1.0 111 km
can be used to sense orientation, coordinate acceleration, 1 0.1 11.1 km
vibration, shock, and falling in a resistive medium. Micro
Machined accelerometers are increasingly present in portable 2 0.01 1.1 km
electronic devices and video game controllers, to detect the 3 0.001 110 m
position of the device or provide for game input. In our
system, accelerometer plays an important role in detecting the
4 0.0001 11 m
potholes and bumps. Characterization of potholes and bumps in 5 0.00001 1.1 m
the roads can be done using the readings of the accelerometer. 6 0.000001 0.11 m
Sudden change in the accelerometer values will indicate the
presence of the pothole in that location. 7 0.0000001 11 mm
Deployment of mobiles in the vehicle is the major 8 0.00000001 1.1 mm
challenge we faced. When the mobile phone is deployed on the
3) GPS Speedometer
vehicle there are many things that should be taken into
Speed is the measurement of distance over time. But a car
consideration to get the accurate reading. The accelerometer
speedometer doesn’t actually measure how fast you travel from
sensor had 3 axis values. Fig. 4 shows the alignment of these 3
Point A to Point B. Car speedos usually work by measuring
axis of accelerometer in the mobile phone.
rotation of the car’s drive shaft, axle or wheel. They then use
some basic math to extrapolate that rotation and determine how
fast you are travelling. It’s a very similar concept to a bicycle
speedometer. However, if the diameter of the wheel/tire alters,
the extrapolation calculation will be incorrect. For example, the
diameter will increase if you put new tires on the car or
increase the tire pressure. This means that, for each revolution
of the wheel, the car is travelling further, meaning your speed
is greater.
Satellite navigation units (either portable or integrated into
the car) calculate your car’s speed by measuring actual distance
Fig. 4. Accelerometer alignment in mobile phone travelled over time using GPS satellite tracking. They
2) GPS Sensor repeatedly locate your exact position on earth via satellite and
A GPS sensor was added to the data acquisition system to calculate how far you have travelled, then divide by the time it
record the location of defects and the speed of the vehicle. If took for you to travel that distance. The accuracy of this
the accelerometer sensor detects the potholes or the bumps in method is determined by satellite signal quality and is
the road, then immediately the Geo-location and speed of the unaffected by your car’s tires.
vehicle has to be calculated from the GPS sensors. The Geo- Speed= Distance/Time
location data will contains the coordinates of the potholes or
bumps. The speed of the vehicle will be used for analyzing the B. Data Processing
data for efficient working of our system. This can be said as the backbone of the system. This
GPS accuracy in our deployment is important if potholes subsystem collects the data from different vehicles and Co-
are to be properly located and multiple detections combined to ordinates the data. It is a web based application. It acts as an
report a single pothole. Accuracy is the tendency of your interface between the user and the system. Central server
measurements to agree with the true values. Precision is the accepts the data send by the application in the URL format. It
degree to which your measurements pin down an actual value. evaluates the data before storing it in the database. These data
2017 International Conference on Innovations in information Embedded and Communication Systems (ICIIECS)

consists of co-ordinate values of the potholes, speed of the REFERENCES

vehicle and the accelerometer difference value. The analysis of [1] Yulu Luke Chen et al., “Inexpensive Multimodal Sensor Fusion
the captured data is done at Server Side and it is used to derive System for Autonomous Data Acquisition of Road Surface
the conclusion from analyzed data i.e. whether road is safe to Conditions” IEEE SENSORS JOURNAL, VOL. 16, NO. 21,
journey or not is carried out and Store it on the web server. NOVEMBER 1, 2016, pp. 7731-7743.
[2] Rajeshwari Madli et al., “Automatic Detection and Notification
C. Plotting the Data of Potholes and Humps on Roads to Aid Drivers”
This subsystem is used to share the stored data in the [3] Juan Martin Raya Bahena et al., “Speed booms detection for a
central server to everyone. It is a web application to display the ground vehicle with computer vision” Col. Lindavista, Del.
condition of the road. This subsystem displays the details of the Gustavo A. Madero, México, D.F. C.P. 07738 MÉXICO.
road as per the location data present in the central server. In [4] Jakob Eriksson, Lewis Girod, Bret Hull, Ryan Newton, Samuel
this method we can plot the markers in the map for every Madden, Hari Balakrishnan, “The Pothole Patrol: Using a
latitude and longitude values present in the server. We are Mobile Sensor Network for Road Surface Monitoring” The
using different markers to differentiate the severity of the road Sixth Annual International conference on Mobile Systems,
condition. These severity measures are done at the server side Applications and Services (MobiSys 2008), Breckenridge,
along with the clustering. If any of the clusters have more U.S.A., June 2008.
values, then we can mark this location as severely damaged [5] Chen et al., “RCM-TAGPS”.
road. We Plot the road conditions based on the surveillance [6] Mednis, A., Strazdins, G., Zviedris, R., Kanonirs, G., Selavo,
values and the analyzed values from the data module. Fig. 5 “Real time pothole detection using Android smartphones with
shows the example values plotted in the map. accelerometers”, 7th IEEE International Conference on
Distributed Computing in Sensor Systems and Workshops
(DCOSS’11) (27-29 Jun. 2011, Barcelona, Spain), IEEE, 1–6.
[7] Bhoraskar, R et al., “Wolverine: Traffic and Road Condition
estimation using Smartphone Sensors”.
[8] Mohan, P., Padmanabhan, V. N. and Ramjee R.; ”Nericell: rich
monitoring of road and traffic conditions using mobile
smartphones” 6th ACM conference on Embedded network
sensor systems, SenSys ’08, pp 323–336.
[9] J. Karuppuswamy, V. Selvaraj, M. M. Ganesh, and E. L. Hall;
“Detection and avoidance of simulated potholes in autonomous
vehicle navigation in an unstructured environment” Intelligent
Robots and Computer Vision XIX:Algorithms, Techniques, and
Active Vision: 70–80, 2000.
Fig. 5. Road condition displayed in the map [10]Sauvik Das -Georgia Institute of Technology, LaToya Green;
“Detecting User Activities using the Accelerometer on Android
IX. CONCLUSION Smartphones‫”ۅ‬, University of Houston, Beatrice Perez
University of Puerto Rico, Mayaguez, Michael Murphy Franklin
Roads has to be monitored continuously for maintaining the W. Olin College of Engineering, 2010.
roads to avoid the inconvenience of the road users. The result
of this activity led to the conclusion that the possibility of
determining the road condition using mobile sensors. The main
aim of this research activity is to find out the location of
potholes and bumps in the road using the mobile sensors. It is a
cost efficient system for maintaining the road condition.
Smartphone sensors are having more importance in this
research, as they are cost effective and increase scalability.