Design and Fabrication of In-Pipe Inspection Robot for Crack Analysis and Detection Using OpenCv and Visual Odometry

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 10 Issue: 07 | July 2023 www.irjet.net p-ISSN: 2395-0072
© 2023, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 353
Design and Fabrication of In-Pipe Inspection Robot for Crack Analysis
and Detection Using OpenCv and Visual Odometry
Alwin Tony1, Adil Noushad2, Aquilas Ninan3, Alwin A V4, Naveen Prakash K V5
1,2,3,4 UG Scholar, Department of Mechanical Engineering, Toc H Institute Of Science and Technology, Ernakulum,
India
5Assistant Professor, Dept. of Mechanical Engineering Toc H Institute Of Science and Technology, Ernakulum, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This project focuses on the design and
enhancement of electronics controlsystemsforpipeinspection
robots. Currently, these robots are primarily manufactured
offshore, resulting in high costs and inadequate support
during malfunctions, leading to environmental mishaps. The
objective is to leverage existing mechanical platforms and
develop a reliable, robust, and cost-effectivein-pipeinspection
robot. Existing pipe inspectionrobotmodels worldwideexhibit
significant performancevariations. Manysystemshavelimited
applicability, lack reliability, and heavily rely on user
experience. This project aims to overcome these challenges by
developing an innovativerobotcapableofnavigatingpipelines
and accurately detecting crack propagation. The robot must
operate reliably in confined environments and provide
technicians with digital video feedback. It will incorporateoff-
the-shelf components, inexpensive materials, and allow for
potential on-site repairs, emphasizing traditional mechanical
engineering principles. It enables the robot to autonomously
distinguish between cracks and other voids, eliminating the
need for additional human intervention. The proposed self-
initiating distinction between cracks and voids streamlines
inspections, minimizinghumanerrorandenhancingefficiency.
Design of a six-legged 120 degree apart symmetric structure
for the robot for easy manoeuvrability equipped with wifi
module, Microcontroller, and set of cameras for computer
vision. Utilises OpenCv and haar Cascading techniques and
python programming to detect crack from live video feedback
and shows the output to user outside. Absolute distance isalso
measured using visual odometry wherein change in pixels of
multiple images is calculated to distance.
Key Words: Pipe Inspection Robot, OpenCv, Haar
Cascading, Visual Odometry, Crack analysis,
Manoeuvrable Robot
1. INTRODUCTION
Pipes are used in many industries to transport variouskinds
of fluids and gases, many of these pipes are long and
inaccessible to technicians and workers, ensuring proper
passage of the fluids and maintenance of the pipes are an
important parameter for the clientand enduser.Thisproject
specifically focuses on pipes used in continuous long
transmission of water and other fluids, pipes used in
divisional transportation of water. Cracksandwaterleakage
are a major concern for regional and national water
authorities, as damages to water pipes are prone due to
hindrance and negligence caused from general public, road
and transport authority because of their subsequent road
maintenance and building. The in-pipe robots consists a
mechanism of camera and sensors to identify these crack
propagation signatures and return a live feed monitoring
with Program based differentiation to a technician for
further facilitation of evaluation. Pipe transmission systems
deteriorate progressively over time. Pipelines are one of the
safest ways to transport water and other liquids. The bigger
part of pipeline transmission systems are accessible by
existing inspection tools or human intervention but this is
limited to the section which are normally at the exterior or
joints of the pipe. Corrosion or cracks does not generate
largely in this areas, hence the requirement of a smaller
inspection device increases. The industry is keen on
receiving newer innovations that allows inspection without
interrupting the routine operations. It is understood that
reliable and accurate information of inspection can only be
found only by direct pipe wall access. The robot (PIR) with
an adaptable design may help in imbibing with the interior
environment of the pipe, especially with respect to the
diameter of the pipe, with high maneuverability and ability
to operate under hostile conditions. For PIR we are using a
six wheeled concept since it is the easiest, most energy
efficient and has the potential to travel long range.
1.1 Review of Literature
[1]2010, Manabo ONO, et al., discusses the possibilitiesof an
Earthworm like robot that may be used for inspections of
long pipelines. It uses a design similar to that of an
Earthworm which uses three rubber bellows as pneumatic
actuators. [2]2012, MD Rashid Ashraf, et al., presents the
procedures undertaken during the design and development
of a small size pipe-in robot for the purpose of boiler head
inspection. This robot used the method of visual inspection
using a camera probe that shall be later looked in to for the
carrying out of the procedure. [3]2012, M F Yousuff, et al.,
discusses about a robot that can be utilized for the
inspection and movement along an air-condition pipeline. It
details the different parameters that are to be met while
designing such a robot. [4]2012, Dongwoo Lee, et al.,
discusses about a methodology for navigating a pipeline
using two mechanisms known as; Adaptable Quad Arm

Mechanisms (AQAM) and the Swivel Hand Mechanism
(SHM). These mechanisms help in bypassing the earlier
difficulties faced by in pipe robots in traversing through a
pipeline. [6]2014, Majid M Moghadam, et al., presents an
apparatus that can be used to traverse a pipeline effectively
by using three independent rubber track units that are 120
degree apart symmetrically. The paper also goes into detail
with regards to the force analysis that was done on the
apparatus. The design phase is clearly mentioned and the
application of each part is noted systematically. [9]2018,
Sami Salama Hussein Hajjaj, et al., The paper mentions in
detail about three possible methods they may use visual
inspection using camera probes, assessment of corrosion
coupons, or pipe Inspection using in line inspection devices
for detecting places of pitting or metal loss. The design is a
caterpillar design for in-pipe inspection. [10]2022, Kaned
Thung-Od, et al., discussed about is one that can be used in
ferromagnetic pipelines that are used to transport Oil, Gas
etc. These robots have both vertical and horizontal mobility
options. The design is in the shape of a train with three
sections. Sensors were provided for measuring various data
from within the pipe.
2. PROJECT METHODOLOGY
1. Designing of pipe inspection robot, here the design of
pipe inspection robot was implemented using NX6.0,
Autodesk Fusion 360 and Autodesk Tinker Hub.
2. Development of pipe inspection robot, here Fusion 360
and Tinker hub was used for the simulation before actual
building of the prototype.
3. Fabrication of Model by means of manufacturing
equipments. This robot consists of six legged wall
caterpillars.
4. Creation of a camera web server in accordance with
arduino microprocessor and testing of initiation of
feedback from ESP32 Camera 1 & Camera 2.
5. Creation and testing of an open CV based python (PY
Charm) interfaced program that will render the live
feedback from ESP32 (1&2) and will use the contours
from the library to differentiate the cracks from the
feedback.
6. Assemble all the required fabricated parts.Thecomplete
assembly of six legs, six wheels, translation element
between the first and second enclosure as well as centre
frame was made using Galvanized Iron fixtures.
7. Relay Motors andDriverofL298N Specificationwasused
in facilitating generic backwardandforwardmovement of
the robot.
8. Fixing of ESP32 cameras, microprocessor and Relay
motors. We will be selecting 6-DC motors.
9. Making it to crawl inside a pipe by means of remote
controller to capture the video image in a screen (Live
video feedback). Evaluation of Visual Odometry and
determining the scaling factor requiredfortheprogramto
convert multiple pixels from image to distance by
cascading.
10. Use of HAAR Cascading to feed in the crack propagation
simulative images and return the feedback.
3. DESIGN OF PIPE INSPECTION ROBOT
3.1 Design Constraints
Size and Dimension: The robot should have a compact size
and dimensions to fit and navigate through pipes of varying
diameters and configurations.
Maneuverability: The robot should be able to navigate
through tight spaces, bends, and obstacles within the pipe
network.
Robustness and Durability: The robot should be built with
materials and components that can withstand harsh and
corrosive environments found in pipes, ensuring long-term
reliability.
Power and Energy Efficiency: The robot should be designed
to operate using efficient powersources,suchasbatteries,to
provide sufficient runtime for inspections without frequent
recharging or replacement.
Sensing and Detection Capabilities: The robot should
incorporate sensors and detection systems capable of
accurately identifying cracks, leaks, blockages, and other
pipe abnormalities.
Communication and Data Transmission:Therobotshouldbe
equipped with reliable communication systems to transmit
inspection data, video feedback, and control signals in real-
time to operators or monitoring systems outside the pipes.
3.2 Design Specifications
To complete the project it is very crucial to have the
specifications before the start of the initial development of
the project. This will ensure the conformity with the set of
standards. Some of the specifications are:-
1. Video streaming 1440p
2. Vertical movement within the pipe (Max. crawl of
10km/h)
3. Wireless control (up to 1200 m)
4. Galvanized Iron body with aluminium endowments
5. 48 Megapixel camera with wideangleapproachsensor for
distance ranging
6. Remote Control using pycharm integrationand remote XY
for elongated control of the robot

7. Rechargeable Power suit, which is waterproof to run the
movement.
3.3 Design of Parts and Assembly
The pipe inspection robot consists of a mechanism of six-
legged 120 – degree apart rotor based traction tire set-up
which is docked into multiple transition element which is
welded into the main central frame, additional pushelement
with spring loaded risers are inserted to eliminate the
change arising within external walls of pipe and thetiresub-
clearance.
Dimensions are L=200mm, H=80mm diameter, and inner
diameter = 17mm, Thickness of extended part= 8mm. there
are total of six extended parts on center frame on 70 mm
from one side and another 3 on 170mm from one end and
have a hole of diameter 6mm.
Fig -1: Central Frame of the Robot using Autodesk Fusion
360
Fig -2: Overall Dislocated assembly of Design (Fusion 360)
3.4 Power Calculation
3.4.1 Assumptions
1. The force required for movement is equal to the
weight of the robot.
2. The weight of the robot is 0.716kg.
3. Surface of the pipe is very smooth.
4. Maximum speed of the robot is 1m/s.
5. The efficiency is 80%.
3.4.2 Calculations
If the speed of the robot is 1 m/second and the robot weighs
0.716 kg, we can calculate the power required for the motor
using the formula:
Power = (Force × Velocity) / Efficiency
First, let's calculate the force required for movement. Since
the robot is on a flat surface, we can assume the force
required is equal to the weight of the robot:
Force = Weight = 0.716 kg × 9.8 m/s^2 (acceleration due to
gravity) = 6.8 N
Assuming an efficiency of 80% or 0.8, we can substitute
these values into the formula:
Power = (6.8 N × 1 m/s) / 0.8 = 8.5 W
Considering the factor of safety we take 10 watts DC geared
motor.
4. DETAILED DESCRIPTION OF PARTS USED
4.1 Arduino Uno:
Microcontroller: The Arduino Uno is built around the
ATmega328P microcontroller, which is an 8-bit AVR
microcontroller.
Clock Speed: The ATmega328P runs at a clock speed of 16
MHz
Digital I/O Pins: The Arduino Uno has a total of 14 digital
input/output pins. Among them, 6 can be used as PWM
(Pulse Width Modulation) outputs.
Analog Inputs: It has 6 analog input pins, labeled A0 to A5,
which can also function as digital I/O pins.
Programming: The Arduino Uno can be programmed using
the Arduino programming language, which is a simplified
version of C++.
Power Supply: The Uno can be powered using a USB
connection, a DC power jack (7-12V), or by supplying 5V
directly to the 5V pin or 3.3V to the 3.3V pin.
Memory: The ATmega328P on the Uno has 32KB of flash
memory for program storage, 2KB of SRAM, and 1KB of
EEPROM for data storage.

Fig- 3:- Graphical Schematic of Arduino UNO
Microcontroller
The Arduino Uno operates by following a specific sequence
of operations. It starts with a power supply provided
through a USB connection, DC power jack, or direct voltage
input. Once powered, the microcontroller runs a bootloader
program stored in its memory, allowing for easy uploading
of sketches. After initialization,theArduinoUnoexecutesthe
setup () function where variables are initialized and pin
modes are configured. The main part of the program is the
loop () function, which is repeatedly executed. Here,youcan
implement project functionality, read sensor inputs, and
control actuators. The board supports interrupts for
immediate response to external events.
4.2 ESP32 CAM
The ESP32CAM is a development board that integrates the
ESP32-S module and a camera sensor.Itcombinesthe power
of the ESP32 microcontroller and the capability of capturing
images and video using a camera module. Here are some
details about the ESP32CAM:ESP32 Microcontroller: The
ESP32CAM is based on the ESP32-WROOM-32 module,
which features a dual-core ESP32 microcontroller. The
ESP32 is a powerful and versatilemicrocontrollerwith built-
in Wi-Fi and Bluetooth capabilities. Camera Sensor: The
ESP32CAM includes a camera sensor module that allows for
capturing images and video. The camera module is typically
based on the OV2640 or OV7670 sensor, offering different
resolutions and capabilities. Flash Memory: The ESP32CAM
board is equipped with 4MB of flash memory, which is used
for storing the firmware and program code.
Fig 4:- Graphical Schematic of ESP32 Camera and its
interfaces
4.3 ESP 8266 Wifi Module
The ESP8266 Wi-Fi module is a popularandaffordable Wi-Fi
module that has gained significant popularity among
developers and hobbyists for its capability to add Wi-Fi
connectivity to various electronic projects. Here are some
details about the ESP8266 Wi-Fi module: Microcontroller:
The ESP8266 module is itself a self-contained
microcontroller system-on-chip (SoC) with integratedWi-Fi
capabilities. It features a 32-bit ten silica Xtensa LX106
microcontrollerunit(MCU).Wi-FiConnectivity: TheESP8266
module supports Wi-Fi connectivity, allowing devices to
connect to local networks and the internet. It supports the
802.11b/g/n Wi-Fi standards, providing reliable wireless
communication. Flash Memory: The ESP8266 module
typically comes with onboard flash memory for storing
firmware, program code, and data. The available flash
memory varies across different models and versions of the
module. GPIO Pins: The ESP8266 moduleprovidesa number
of general-purpose input/output (GPIO) pins that can be
used for digital input/output, analog input, and various
communication protocols such as UART, I2C, and SPI. The
exact number of GPIO pins varies depending on the specific
module variant.
4.4 Channel Relay Module
Fig 5:- Connection between arduino microcontroller and
Channel relay module
A 5V 2-channel relay module is an electronic module that
allows you to control high-power devices using low-power
Signals from a microcontroller or other control circuit.

Here are some details about a typical 5V 2-channel relay
module: Number of Channels: A 5V 2-channel relay module
has two separate relay channels. Each channel can control a
different electrical load independently. Relay Type: The
relays used in the module are usually electromechanical
relays. These relays consist of a coil and one or more sets of
contacts that can switch high currents or voltages. Input
Voltage: The module is designed to be driven by a 5V DC
power supply. It is compatible with most microcontrollers
and development boards that operate at 5V logic levels.
Control Signal: The module typically uses a low-level control
signal (e.g., logic low) to activate the relays. This control
signal is provided by a microcontroller or any other digital
output device. RelayContacts:Eachrelaychannel usuallyhas
a common (COM), normally opens (NO),andnormallyclosed
(NC) contact. The COM contact is connected to the power
supply or the load, while the NO and NC contacts are used to
switch the load on and off. Opt-isolation: Some relay
modules include opt-isolation circuitry, which provides
electrical isolation between the control circuit and the high-
power side (relay contacts). This helps protect the control
circuit from voltage spikes or disturbances caused by the
switching of inductive loads.
5. OVERVIEW OF PROGRAMS CREATED FOR
ANALYSIS
5.1 RemoteXY Code for Movement of Robot
#include <RemoteXY>
#define REMOTEXY_SERIAL Serial
#define REMOTEXY_SERIAL_SPEED 115200
#define REMOTEXY_WIFI_SSID "RemoteXY"
#define REMOTEXY_WIFI_PASSWORD "12345678"
#define REMOTEXY_SERVER_PORT 6377
#pram pack (push, 1)
uint8_t RemoteXY_CONF[] = // 44 bytes
{ 255,2,0,0,0,37,0,16,31,1,10,48,24,24,15,15,4,26,31,79,
78,0,31,79,70,70,0,10,48,24,75,15,15,4,26,31,79,78,0,31,
79, 70,70,0 };
struct {
uint8_t pushSwitch_1; // =1 if state is ON, else =0
uint8_t pushSwitch_2; // =1 if state is ON, else =0
uint8_t connect_flag; // =1 if wire connected, else =0
} RemoteXY;
#pragma pack (pop)
#define PIN_PUSHSWITCH_1 4
#define PIN_PUSHSWITCH_2 5
void setup()
{
RemoteXY_Init ();
pinMode (PIN_PUSHSWITCH_1, OUTPUT);
pinMode (PIN_PUSHSWITCH_2, OUTPUT);
}
Void loop ()
{
RemoteXY_Handler ();
digitalWrite(PIN_PUSHSWITCH_1,
(RemoteXY.pushSwitch_1==0)?LOW:HIGH);
digitalWrite(PIN_PUSHSWITCH_2,
(RemoteXY.pushSwitch_2==0)?LOW:HIGH);
}
5.2 OpenCv Python Program for Crack Detection
import cv2
import numpy as np
import urllib.request
stream_url = 'http:/192.168.12.269/video'
threshold1 = 30
threshold2 = 100
def process_frame(frame):
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
edges = cv2.Canny(gray, threshold1, threshold2)
cv2.imshow('Crack Detection', edges)
def main():
stream = urllib.request.urlopen(stream_url)
bytes = bytes()
while True:
# Read a frame from the video stream
bytes += stream.read(1024)
a = bytes.find(b'xffxd8')
b = bytes.find(b'xffxd9')
if a != -1 and b != -1:
jpg = bytes[a:b+2]
bytes = bytes[b+2:]
frame cv2.imdecode(np.frombuffer(jpg,
dtype=np.uint8), cv2.IMREAD_COLOR)
process_frame(frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
stream.release()
cv2.destroyAllWindows()
if __name__ == '__main__':
main()
5.3 HAAR Cascading Program fordistancemeasurement
using visual odometry
url = http://102.167.652.29:81/stream
cap = cv2.VideoCapture(url)
prev_frame = None
dist_traveled_pixels = 0
dist_traveled_centimeters = 0
prev_points = None
scaling_factor = 0.1235
lk_params = dict(winSize=(15, 15), maxLevel=2,
criteria=(cv2.TERM_CRITERIA_EPS |
cv2.TERM_CRITERIA_COUNT, 10, 0.03))
displacement_threshold = 2.0 # Adjust this value based on
the expected motion magnitude
while True:
ret, frame = cap.read()
if not ret:
break
gray = cv2.cvColor(frame, cv2.COLOR_BGR2GRAY)

if prev_frame is not None:
if prev_points is not None:
next_points, status, _ =
cv2.calcOpticalFlowPyrLK(prev_frame, gray, prev_points,
None, **lk_params)
good_old = prev_points[status == 1]
good_new = next_points[status == 1]
if len(good_old) > 0 and len(good_new) > 0:
displacements = good_new - good_old
displacement_magnitudes = np.linalg.norm(displacements,
axis=1)
average_displacement =
np.mean(displacement_magnitudes)
if average_displacement > displacement_threshold:
dist_traveled_pixels += average_displacement
dist_traveled_centimeters = dist_traveled_pixels *
scaling_factor
prev_points = cv2.goodFeaturesToTrack(gray, 100, 0.3, 7)
prev_frame = gray.copy()t
text = f"Distance Traveled: {dist_traveled_centimeters:.2f}
centimeters"
cv2.putText(frame, text, (10, frame.shape[0] - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 1,
cv2.LINE_AA)
cv2.imshow("Frame", frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cap.release()
cv2.destroyAllWindows()
6 DISTANCE MEASUREMENT USING VISUAL
ODOMETRY
Visual odometry, combined with OpenCv and Python
programming, offers a powerful approach for detecting the
location of a pipe inspection robot within a pipe network.
Visual odometry relies on the analysis of visual data
captured by the robot's onboard camera to estimate its
position and orientation within the pipe. By leveraging
OpenCv, a versatile computer vision library, and Python
programming, the robot can extract and process visual
features, track their movements, and calculate its location.
The process begins with feature detection, where OpenCv
feature extraction algorithmsidentifydistinctvisual features
such as corners or edges within the captured images. These
features act as reference points for subsequent tracking and
location estimation. The features are then matched across
consecutive frames, utilizing OpenCv feature matching
algorithms, to establish correspondences and track their
motion. Python programming is utilized to implement the
visual odometry pipeline.Thetrackedfeaturesareprocessed
using various techniques such as motion estimation, pose
estimation, and triangulation. To ensure accurate location
detection; camera calibration is a critical step. By accurately
calibrating the camera, distortion and other camera-specific
factors are taken into account, improving the accuracyofthe
location estimation. Real-time locationdetectionisachieved
by implementing the visual odometry pipeline efficiently.
Python's versatility and OpenCv optimized functions allow
for fast processing and tracking of visual features, enabling
the robot to estimate its location in real-time during the
inspection process. It is important to note that visual
odometry for pipe inspection robots using OpenCv and
Python has its limitations. Factors such as lighting
conditions, Occlusions and dynamic environments within
the pipes can impact the accuracy of feature detection and
tracking. Therefore, careful consideration must be given to
address these challenges and developrobustalgorithmsthat
can handle various pipe inspection scenarios.
7 CRACK DETECTION AND RESULTS
The pipeline undergoes testing by intentionally introducing
different types of cracks with varying sizes and surface
areas. These cracks include major, minor, and elongated
ones. The propagation of these cracks can be a significant
cause of intermittent pipeline failures,leadingtodisruptions
in product transmission. To effectively inspect the pipe's
interior, a pipe inspection robot equipped with an ESP-32
Camera is utilized. It captures detailedfootageandtransmits
real-time feedback via a live IP address. This IP address is
then utilized in a Python program that invokes OpenCv for
further analysis. OpenCv accesses the live video feedback
and processes it to identify irregular shapes and contours.
OpenCv enhances the analysis by applying a mask to
differentiate the crack from the rest of the system. This
masking technique helps in isolating and highlighting the
crack, making it easily identifiable. The crack is distinctly
marked with a red contour, allowing personnel to
effortlessly determine its position within the extensive
pipeline network. By employing this comprehensive
inspection process, pipeline operators can proactively
address cracks and potential failures, ensuring the
uninterrupted transmission of productsandmaintaining the
integrity of the pipeline system.
Fig 6: Crack induced on pipe with marked edges to
highlight the irregularity

Fig 8:- Crack detection using OpenCv (from test cracks)
8 CONCLUSIONS
In conclusion, the project work on a pipeline inspection
robot utilizing Arduino Uno, RemoteXY, relay module, DC
motors, RemoteXY app for movement control, ESP32-CAM,
and an OpenCv Python program for crack detection offers a
comprehensive solution for efficient and accurate pipeline
inspection. The integration of Arduino Uno provides a
reliable microcontroller platform for controlling the robot's
movements and coordinating various components. The
RemoteXY library enables easy communication betweenthe
robot and the RemoteXY app, allowing intuitive control of
the robot's forward and backward movement. The relay
module plays a crucial role in controlling the DC motors,
facilitating the robot's locomotion. The inclusion of the
ESP32-CAM module enhances the robot's capabilities by
enabling live video streaming from the pipeline.The OpenCv
Python program leverages thepowerofcomputervisionand
image processing techniques, specifically canny edge
detection, to detect cracks in real-time. This provides
valuable insights into the condition of the pipeline andhelps
identify potential issues for timely maintenance. By
combining these technologies, the pipeline inspection robot
offers several advantages. It enhances efficiency by
automating the inspection process, reducing the need for
manual intervention and saving time. The accuracy of crack
detection is improved through the utilization of computer
vision algorithms, providing detailed visual information for
analysis and decision-making , For instance, implementing
autonomous navigation, advancedsensingtechnologies, and
data analytics capabilities can further enhance the robot's
performance and provide more comprehensiveinsightsinto
the pipeline's condition. The pipeline inspection robot
project utilizing Arduino Uno, RemoteXY, relay module, DC
motors, RemoteXY app, ESP32-CAM, and OpenCv Python
program showcases a practical and effective solution for
pipeline inspection. It combines hardware, software, and
computer vision techniques to improve efficiency, accuracy,
and safety in pipeline maintenance and management. The
project serves as a foundation for further advancementsand
research in the field of pipeline inspection robotics.
REFERENCES
[1]M.F. Yusoff, B.S.K.K. Ibrahim, H. Hamzah, H.A. Kadir
Development of Air Conditional Route Wireless Inspection
Robot International Symposium on Robotics and Intelligent
Sensors (2012), pp. 874-880 (IRIS 2012)
[2] Md Raziq Asyraf Md Zin, Khairul SallehMohamedSahari,
Juniza Md Saad, Adzly Anuar, and Abd Talip Zulkarnain,
Development of a Low Cost Small Sized In-Pipe
RobotInternational Symposium on Robotics and Intelligent
Sensors 2012 (IRIS 2012), pp 1469-1475.
[3] DongwooLee, JungwanPark, DongjunHyun, GyungHwan
Yook, Hyun-seok Yang Novel mechanisms and simple
locomotion strategies for an in-pipe robot that can inspect
various pipe type Mechanism and Machine
Theory, 56 (2012), pp. 52-68
[4] MajidM.Moghaddam, Mohammadreza Arbabtafi, A. Hadi
In-pipe Inspection Crawler Adaptable to The Pipe Interior
Diameter International Journal of Robotics and
Automation, 26 (2) (2011), pp. 135-145
[5] Megumi Ikeuchi, Taro Nakamura, and Dai Matsubara,
Development of an In-pipe Inspection Robot for Narrow
Pipes and Elbows using Pneumatic Artificial Muscles, 2012
IEEE/RSJ International ConferenceonIntelligentRobots and
Systems October 7-12, 2012. Vilamoura, Algarve, Portugal,
pp 926-931.
[6] Yazdekhasti, S.; Piratla, K.R.; Sorber, J.; Atamturktur, S.;
Khan, A.; Shukla, H. Sustainability analysis of a leakage-
monitoring techniqueforwaterpipelinenetworks. J. Pipeline
Syst. Eng. Pract. 2020, 11, 04019052
[7]Rodríguez-Olivares, N.A.; Cruz-Cruz, J.V.; Gómez-
Hernández, A.; Hernández-Alvarado, R.; Nava-Balanzar, L.;
Salgado-Jiménez, T.; Soto-Cajiga, J.A. Improvement of
Fig 7:- Robot inspecting the interior walls of the Test Pipe

ultrasonic pulse generator for automatic pipeline
inspection. Sensors 2018, 18, 2950
[8] Henry, R.; Chablat, D.; Porez, M.; Boyer, F.; Kanaan, D.
Multi-objective design optimizationofthelegmechanism for
a piping inspection robot.In ProceedingsoftheInternational
Design Engineering Technical Conferences and Computers
and InformationinEngineeringConference,Buffalo,NY,USA,
17 August 2014; p. V05AT08A001.
[9] Singh, W.S.; Rao, B.; Sasi, B.; Vaidyanathan, S.; Jayakumar,
T.; Raj, B. Giant Magneto-Resistive (GMR) Sensors for Non-
Destructive Detection of Magnetic Flux Leakage from Sub-
Surface Defects in Steels. In Proceedings of the International
Conference on Sensors and Related Networks (SENNET-
2009), VIT University, Vellore, India, 1 December 2009;
[10] Kreutzbruck, M.; Neubauer, A.; Pelkner, M.; Reimund,V.
Adapted gmr array used in magnetic flux leakageinspection.
In Proceedings of the 18th World Conference on
Nondestructive Testing, Durban, South Africa, 16–20 April
2012;

Design and Fabrication of In-Pipe Inspection Robot for Crack Analysis and Detection Using OpenCv and Visual Odometry

More Related Content

Similar to Design and Fabrication of In-Pipe Inspection Robot for Crack Analysis and Detection Using OpenCv and Visual Odometry (20)

More from IRJET Journal (20)

Recently uploaded (20)

Design and Fabrication of In-Pipe Inspection Robot for Crack Analysis and Detection Using OpenCv and Visual Odometry