UNDER THE GUIDANCE OF

Mr.S.RAMAKRISHNA
(ASSOCIATE PROFESSOR,dept of ECE)

Prepared And Presented By

SAURABH SINGH
(1DT14EC071)
 Abstract
 Introduction
 Proposed System Overview
 Block Diagram of Proposed System
 Accelerometer
 Pre-Processing
 Abnormal Event Detection
 Event Classification
 Event Severity Estimation
 Conclusion And Future Work
 References
In India, road transportation dominates all other means of transportation
mechanisms. Well maintained roadways make the travel smooth and
comfortable for the passengers. Most of the time road irregularities such
as potholes and humps create disturbances to easy travel and cause
major damages to vehicles. Detecting road potholes and road roughness
levels is a key to road condition monitoring, which impacts on transport
safety and driving comfort. This paper proposes a method that aims to
monitor road surfaces, detect road potholes and humps and predict their
severity by analysing the vertical vibration signals produced by the
vehicle while it moves. The proposed system uses smartphone
accelerometer to capture the vehicle vibrations in which Z-axis reading
corresponds to the vehicle vertical vibrations. Gaussian model based
mining algorithm is proposed for the abnormal event detection, X-Z ratio
filtering is applied for event classification as pothole or hump. Severity
estimation algorithm is proposed, which makes use of the relation
between vertical acceleration and relative vertical displacement of the
vehicle.
 Current road surface monitoring uses human effort to check the
condition and quality of the road which makes the process more
time consuming and less efficient.
 An automatic system which is capable of detecting road damages
without any human effort may completely transform the current road
transportation system efficiency.
 Identification of pavement distress such as potholes and humps
not only helps drivers to avoid accidents or vehicle damages, it also
helps authorities to keep track of the road conditions for its better
maintenance.
 In the past decade there were several methods that came into
picture for the detection and identification of such distress on the
roads.
 Even though these approaches try to identify the presence of
pothole and hump, these techniques are very expensive in nature
and it cannot be applied to middle class level vehicles.
(…continues)
With this as our major objective we try to propose a road surface
monitoring technique which detects and estimates the severity of the
irregularity event based on vehicle vibration analysis using smartphone
tri-axial accelerometer.
The proposed road surface monitoring system consist of
Data collection and Pre-processing.
Abnormal event detection from the processed data samples.
Classification of the event into different categories.
Estimation of severity of the detected abnormal event.
 Accelerometer is a sensor, which converts acceleration from motion
(dynamic acceleration) or gravity (static acceleration) to either
analog or digital electrical signals.
 Dynamic acceleration is caused by the actual motion of the sensor
whereas force of gravity is the one which causes static acceleration.
 Smartphone accelerometer sensors sense the acceleration event of
smartphones. Most of the smartphone accelerometer reading
contains three axes whose directions are pre-defined.
 Most of existing accelerometers comes with user friendly
configurations which allows the user to choose a best sampling rate
through experiments.
 Most of the smartphone accelerometer reading contains three axis
whose directions are pre-defined as in below figure.
There are two stages in Pre-Processing
 Segmentation of raw data
• The collected acceleration vibration data is segmented into groups of
n samples.
• This approach allows the system to identify the occurance of multiple
abnormal events.
 Max-Abs Filtering
• Max-Abs filter retains the data samples corresponding to an abnormal
event (Pothole or Hump).
• Max-Abs filter minimizes all other small acceleration spikes from the
segmented data.
 Event detection can be defined as a process of identifying the
significant abnormalities such as potholes and humps on the road.
 The smartphone accelerometer senses the vertical vehicle vibration
by measuring the Z-axis acceleration samples.
 It can be easily found that the vibrations on abnormal road sections
are much greater than that on the normal road surfaces.
 Sudden change in the Z-axis acceleration often leads to an
abnormal event.
 Once we have identified and labelled a data sample as an abnormal
event, we need to decide whether it belongs to a pothole or a hump.
 To do this classification, we make use of X-Z filter.
 X axis acceleration can be used to classify the abnormal event into
a pothole or hump.
 In most of the cases pothole impacts only one side of the vehicle
and it produces a large Z axis peak along with a significant rise in
the X axis reading for a short duration of time.
 Whereas when the event is a hump, it impacts on both sides of the
vehicle and it produces a large Z axis peak accompanied by a
relatively small X axis acceleration.
 After the detection and classification of the abnormal road condition
event, the next function of the system is to estimate the severity
(Depth or Height) of the event.
 The severity of abnormal event for example, the depth of pothole is
assumed to be related to the relative vertical displacement caused
to the vehicle while it passes through the pothole.
 In this paper, we propose the design and implementation of a vehicle
vibration signal processing system for road surface monitoring. A
smartphone in-built accelerometer data is used in which the proposed
algorithms are applied. The experimental results show that the
proposed system can detect and classify the abnormal events such as
pothole and hump from the collected data at any vehicle driving
speed. It can also estimate the severity of the identified event when
the test driving speed falls in the range of 15 to 20km/h. Our system
focuses on monitoring of road abnormalities such as pothole and
hump.
 The future work may include the detection of other irregularities like
Expansion joints, Manhole and Pipeline holes etc. In future the
researchers can also investigate methods to estimate the event
severity at any vehicle driving speed.
[1] Seung ki Ry and Taehyeong Kim, Review and analysis of pothole
detection methods, Journal of emerging trends in computing and
information science, Vol. 5, No. 8 August 2014.
[2] Xiao Ai, Zhen Zhang and C.K Chan, An efficient algorithm for
pothole detection using stereo vision, IEEE international Conference on
Acousitc, Speech ans signal Processing, Pages: 564 - 568, 2003.
[3] Santhosh Hebbar, Rajeshwari Madli, Varaprasad Golla and
Praveenraj Pattar, Automatic Detection and Notification of Potholes
and Humps on Roads to Aid Drivers, IEEE Sensors Journal, Volume.
15, No. 8, August 2015.
[4] Yong Yang , Jinwoo Jang, Andrew W. Smyth and Dave Cavalcanti,
Road Surface Condition Monitoring via Multiple Sensor-Equipped
Vehicles, IEEE Infocom Poster Presentation, Pages: 43 - 44, 2015.
[5] Hamshan Khan and Vinay Rishiwal, Automatic pothole and speed
breaker detection using android system, MIPRO 2016, Pages: 1270
1273, June 3,2015
[6] Mingming Lu, Guang Tan, Kongyang Chen and Jie Wu, CRSM:
Crowd sourcing based Road Surface Monitoring, IEEE International
Conference on High Performance Computing and Communications,
Pages: 2151 2158, 2013.
[7] Lewis Girod, Bret Hull and Jakob Eriksson, The Pothole Patrol:
Using a Mobile Sensor Network for Road Surface Monitoring, The Sixth
Annual International conference on Mobile Systems, Applications and
Services (MobiSys 2008), Breckenridge, U.S.A., June 2008.
[8] Chenglong Liu, Difei Wu,Yuchuan Du and Siyu Li, Application of
Vehicle Mounted Accelerometers to Measure Pavement Roughness,
International Journal of Distributed Sensor Networks, Volume 2016,
Article ID 8413146, pages: 234-241, June 2016.