Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)

IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

Navigation of Mobile Robot with Nav2 and

SLAM using LiDAR
Suwash Shrestha Sudip Parajuli
Department of Electronics and Computer Engineering. Department of Electronics and Computer Engineering.
Institute of Engineering, Purwanchal Campus Dharan Institute of Engineering, Purwanchal Campus Dharan
2025 4th International Conference on Sentiment Analysis and Deep Learning (ICSADL) | 979-8-3315-2392-3/25/$31.00 ©2025 IEEE | DOI: 10.1109/ICSADL65848.2025.10933029

Dharan, Nepal Dharan, Nepal

suwash.077bei@ioepc.edu.np 077bei041@ioepc.edu.np

Pawan Shah Basanta Shrestha

Department of Electronics and Computer Engineering. Department of Electronics and Computer Engineering.
Tribhuwan University, IoE, Eastern Regional Campus Tribhuwan University, IoE, Eastern Regional Campus
Dharan, Nepal Dharan, Nepal
077bei025@ioepc.edu.np 077bei008@ioepc.edu.np

Manoj Kumar Guragai

Department of Electronics and Computer Engineering.
Tribhuwan University, IoE, Eastern Regional Campus
Dharan, Nepal
manoj@ioepc.edu.np

Abstract—This paper dives into the world of autonomous mo- Cartographer and Navigation2 (Nav2), represents a significant
bile robot navigation that achieves seamless integration between step forward in tackling these fundamental challenges [2],
Robot Operating System (ROS2) and microcontroller-based sys- [3], [4]. ROS2, along with frameworks like Cartographer and
tems without relying on standard ROS control packages. Unlike
conventional methods that depend on dedicated ROS hardware Navigation2 (Nav2), marks a significant impact in addressing
integration packages, the system implements a traditional serial the core challenges of mobile robotics. Although many papers
communication that enables direct interfacing with Arduino [5], [6] rely on Visual Simultaneous Localization and Mapping
microcontrollers. Upon successful demonstration, real-time data (vSLAM) for precise navigation, this approach often demands
handling and efficient buffer management on Arduino Mega excessive computational resources, making it less practical
was achieved by overcoming the memory limitations encountered
with Arduino Uno. The implementation includes a custom joint for real-world scenarios. Alternatively, using Lidar with the
state publisher that ensures precise synchronization between Nav2 framework can be a great way to dynamically map
the physical robot and its simulated counterpart along with the environment and enable autonomous navigation. This
the deployment of the Robot Operating System (ROS2) and paper seeks to improve the combined capabilities of ROS2,
the Navigation Stack (Nav2) with Cartographer ROS, using Cartographer, and Nav2 to enhance the performance of a
their capabilities for precise environment mapping, navigation,
and complex path planning in mobile robotics. Our system Four-Wheeled Differential Drive Mobile Robot (4WDDMR)
achieves accurate robot movement through careful calibration in executing intricate tasks related to localization, mapping,
and real-world parameter measurements, incorporating detailed and navigation.
inertial calculations for improved motion control. This approach
improves the precision of autonomous navigation and broadens 2D Light Detection and Ranging (LIDAR) Slam is consid-
the applicability of the robot in various scenarios, ranging ered one of the techniques for real-time mapping of environ-
from disaster management and indoor navigation to warehouse ments with unparalleled accuracy [4]. Precise self-localization
logistics. is of great importance in robotics, especially in autonomous
Index Terms—Autonomous Navigation, ROS2, SLAM, Nav2, navigation systems such as those made using the ROS2 Nav-
LIDAR, Mobile Robotics, Mapping, Localization
igation Framework (Nav2). These robots find great utility in
various industries. Initially, the Unified Robotics Description
I. I NTRODUCTION
Format (URDF) is implemented, serving as the foundational
In the landscape of mobile robotics, achieving precise map- file for the robot, furnishing crucial information on robot joints
ping and localization is necessary to enable autonomous navi- and movements. This facilitates the robot’s comprehension
gation. The Robot Operating System (ROS) stands as the open- of its structure effortlessly and enables effective control for
source robotics community, which was conceived in 2007 precise movement within an environment. Using the differ-
through the efforts of the Willow Garage robotics incubator ential drive equation and the 2D LIDAR Slam, the robot
[1]. The emergence of ROS2, along with frameworks such as can be efficiently localized, enabling seamless navigation.

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 579

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)
IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

The ability of ROS2 to perform multiprocessing for real-time a relatively modular yet flexible framework for developing
performance, coupled with its robust security features [2], robotic systems, consisting of libraries and tools for help in
improves the functionality of the system. perception, planning, control, and many other domains. Of
all the capabilities of ROS2, it definitely adds a lot more
functionality to the existing ROS features. It improves real-
time performance, security, and scalability.
That is why ROS2 would be the right platform for the
project at hand, offering the full support of multi-robot sys-
tems and conforming to the modern hardware and software
ecosystems used in the development of automatic navigation
systems.

C. Navigation2 (Nav2) Framework

However, while the framework is modular and robust, it
requires careful fine tuning for specific applications to achieve
optimal performance. The need for this fine tuning arises
from the diversity of environments and use cases, where
Fig. 1. 4 wheel differential drive mobile robot prototype. the sensor setup, algorithm parameters, and decision making
processes must be tailored to meet the specific demands of the
Fig. 1 illustrates the real robot used in the study. This work application, ensuring high efficiency and reliability in various
also address the limitations of previous work, notably focusing real world scenarios.
on hardware interfacing without reliance on the ros control
package. A custom interface for compatibility with Arduino D. LIDAR-Based Mapping and Localization
Mega was introduced. The main motivation behind this paper
However, one significant challenge with LIDAR-based sys-
is to extend the robust functionality and versatility of existing
tems arises in featureless environments, such as open fields,
ROS2 and NAV2 within robotics applications.
long corridors, or empty rooms, where the absence of dis-
II. L ITERATURE R EVIEW tinctive landmarks or objects makes it difficult for the robot
to maintain accurate localization. In these environments, the
A. Autonomous Navigation and SLAM in Robotics
lack of sufficient features for the LIDAR to differentiate
Robotics has witnessed one of its most excellent break- can lead to problems like drift, loop closure failure, and
throughs in terms of autonomous navigation. With this, robots decreased map consistency. Consequently, additional sensors
can function independently and effectively in dynamic envi- or alternative strategies, such as incorporating visual or inertial
ronments. Some of the popular technologies in this regard are measurements, are often required to enhance performance in
based on Simultaneous Localization and Mapping (SLAM): such feature-poor environments.
SLAM allows robots to build unknown environment maps and
localize themselves at the same time while moving around E. Differential Drive Robots
already explored maps [13]. Many works consider SLAM
Differential drive robots are a common platform for au-
from the different approaches of different types of sensors
tonomous navigation research due to their simplicity and ma-
and components, especially cameras, LIDAR, and ultrasonic
neuverability. These robots use two independently controlled
sensors.
wheels to achieve motion, allowing them to rotate in place
Most conventional methods are based on vSLAM, which
and navigate tight spaces. The kinematics of differential drive
uses a monocular or stereo camera to collect environmental
systems are well-understood, with equations describing the
data. However, vSLAM methods are computationally intensive
relationship between wheel velocities and the robot’s linear
and provide poor performance in low-light or featureless
and angular motion [15]. This makes them an ideal platform
conditions. [14]
for testing and validating navigation algorithms, particularly
In contrast, being based on LIDAR, SLAM is expected
in indoor environments.
to provide higher accuracy and robustness against vSLAM,
especially in structured environments; hence, it is most appro-
F. Challenges in Hardware Integration
priate for applications that require highly accurate mapping
and localization but with almost real-time operation. One of the significant challenges in autonomous robotics
is the integration of hardware components, particularly when
B. Robot Operating System (ROS) and ROS2 using microcontrollers such as the Arduino Mega. While
Introduced in 2007, since then, the Robot Operating Sys- ROS2 offers robust tools for software development, hardware
tem has become a de facto standard within the worldwide interfacing often requires custom solutions, especially when
community of robotics research and development. It provides dealing with real-time data acquisition and control.

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 580

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)
IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

G. Applications of Autonomous Navigation

Autonomous navigation systems have a wide range of
applications, from warehouse logistics and indoor navigation
to disaster management and search-and-rescue operations. In
warehouse settings, autonomous robots can optimize material
handling and inventory management, improving efficiency and
reducing operational costs. [16]
III. T HEORETICAL BACKGROUND
The primary component of autonomous robotics is naviga-
tion, which allows robots to move around their surroundings
safely and effectively to accomplish particular objectives.
It involves several interrelated elements, including mapping,
odometry, path planning, and localization. In robotics, local-
ization is the process of figuring out where a robot is located in
its environment using integrated sensor data from LIDAR and
an encoder motor. Robot motion variations in wheel reading
values from encoder motors are tracked using odometry, which
offers useful feedback for localization and control [7]. When
it comes to storing complicated robot behaviors, such as
navigation methods like goal pursuit and obstacle avoidance,
behavior trees offer a hierarchical decision-making structure.
Autonomous navigation has been transformed in recent
years by the introduction of powerful navigation frameworks Fig. 2. Differential drive representation of robot in Cartesian form.
like Navigation 2 (Nav2). Nav2 is based on the Robot Operat-
ing System 2 (ROS 2), which provides a scalable and modular
framework for navigation activities such as path planning, XB current, YB current be the position of robot center before
control, localization, and map creation [8], [9]. Because of ∆t time,
its adaptability, it can be integrated with a variety of sensors ω be the angular velocity of the robot,
and algorithms, which makes it appropriate for a broad range v be the linear velocity of the robot,
of robotic platforms and applications. Autonomous robots can ∆ϕ be the change in orientation in ∆t time
efficiently and precisely navigate complicated surroundings r be the radius of the wheel, then
by combining Nav2’s navigation skills with localization and b be the separation between the two wheels
odometry. Some works on simulation can be found in [10].
A. Differential Drive Equation VR − VL
Angular velocity, ω = (1)
b
The relationship between the robot’s linear and angular
velocities and the velocities of its wheels is expressed by a VR + VL
differential drive equation. It does not need to use additional Linear velocity, v = (2)
2
steering; instead, it uses skid steering [11]. Fig. 2 explains the
current position of the robot and determines the next position
Change in Orientation, ∆ϕ = ω∆t (3)
with the calculation of angular velocity and linear velocity
from the wheel velocity. For the robot to change its orientation, The position of the robot in the plane can be updated as
it must rotate about a point that lies along its common left follows:
and right wheel axes. The point that the robot rotates about is
known as the Instantaneous Center of Curvature (ICC) and R XB new = XB current + v cos(ϕ)∆t (4)
is the signed distance from the ICC to the center of the robot
[17].
Let VL be the velocity of the left wheel, YB new = YB current + v sin(ϕ)∆t (5)
VR be the velocity of the right wheel,
XB , YB be the position of the robot center, ϕnew = ϕcurrent + ∆ϕ (6)
ϕ be the angle between the X-axis and the angular velocity,
∆t be the time difference between the previous and current These equations update the robot’s position and orientation.
state, The new position coordinates (XB new, YB new) and orienta-
XB new, YB new be the new position of robot center after ∆t tion (ϕnew ) are computed based on the current position, linear
time, velocity, the direction of movement, and the time interval [17].

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 581

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)
IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

IV. R ELATED W ORK TABLE I

H ARDWARE AND S OFTWARE C OMPONENTS
With major developments in mapping, localization, and
path planning over the years, autonomous navigation has Component Specification/Software Used
Microcontroller Arduino Mega 2560
become a primary topic in robotics. The community now has Sensors LIDAR
modular and scalable tools to address these issues thanks to the Motors 12V DC Motors with Encoders
development of strong frameworks like the Robot Operating Communication Protocol UART, Data Distribution Service (DDS)
System (ROS) and its successor, ROS2. Software Framework ROS2 Humble
Custom Libraries Motor Control, Sensor Data Acquisition
Due to the widespread availability of reasonably priced Motor Driver Monster Motor Driver (VNH2SP30)
monocular cameras, Simultaneous Localization and Mapping Development Environment Arduino IDE, VS Code, RViz, ROS2
(SLAM) has been the subject of much research, with numer-
ous methods utilizing Visual SLAM (vSLAM). Studies such
as ORB-SLAM [5] and its variations have shown excellent
visual mapping accuracy. However, as mentioned in [6], these
techniques are computationally demanding as well as poor at
recognizing plain paints on walls, which limits their usefulness
in real life.
Because they provide robustness to lighting variations and
high-resolution environmental mapping, LIDAR-based SLAM
algorithms have gained popularity as a viable substitute. Real-
time 2D and 3D mapping has been made possible by the
successful integration of algorithms like Cartographer [4] into
ROS frameworks. According to [8], the application of LIDAR
improves the accuracy of obstacle identification and map
creation, especially in structured settings like warehouses.
A key component for motion control and path planning has
been the ROS Navigation Stack (Nav1). But because of its lim-
Fig. 3. 4WDDMR URDF virtual representation of real robot with above
ited scalability and dependence on single-threaded operation, parameters.
ROS2 developed Navigation2 (Nav2) [3]. By adding multi-
threaded processing, lifecycle management, and improved
security features [2], Nav2 overcomes these drawbacks and necessitates custom solutions for hardware interfacing and data
makes navigation reliable in challenging situations. handling, particularly in managing real-time data acquisition
Research has also focused on custom hardware integration. from sensors and motors.
The ROS control package is used in many studies to interface During the development phase, a buffer full problem was
with the robot hardware [9]. Custom methods, like the Arduino encountered on the Arduino Uno, which limited the system’s
Mega approach suggested in this study, are less frequently ability to handle extensive data streams from the sensors.
investigated but offer insightful information about affordable The Arduino Uno’s limited memory capacity caused frequent
and customized solutions. overflows when processing real-time encoder data, leading to
In order to accomplish real-time navigation with improved significant performance issues. To mitigate this problem, a
performance, this work combines ROS2, Cartographer, and switch was made to the Arduino Mega, which offers increased
Nav2, building on the foundation established by earlier inves- memory and processing power. This change allowed for more
tigations. It also discusses the adaptability of custom hardware effective data management and maintained real-time perfor-
interfaces and the computational efficiency of LIDAR-based mance for the joint state publisher node and other critical
mapping, which help to expand the use of autonomous mobile processes.
robots in a variety of contexts.
B. Custom Joint State Publisher Node
V. P ROPOSED W ORK
To provide real-time data from the encoder motor, a custom
A. Hardware and Software Setup joint state publisher node is developed. This node publishes the
In this work, the system is developed without relying on the state of the robot’s joints, including position and speed, via the
‘ros control’ package, which typically provides the necessary /joint_states topic. This data is critical for simulating
hardware interfacing for ROS-based projects. While ‘ros con- the robot’s movement accurately in RViz, reflecting the actual
trol’ functions work effectively in ROS, the transition to ROS2 robot’s movements in real time. As well as it is also crucial
introduces challenges, especially regarding micro-controller for the navigation and mapping part for information on the
compatibility. Specifically, ‘micro-ROS’, a package designed transformation frame (tf) that helps with the precise control of
to facilitate ROS2 on micro-controllers, is not installable on the robot. Fig. 3 represents the robot made up of the above-
the Arduino Mega, which is used in this work. This limitation mentioned parameters, which is placed in its Unified Robotics

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 582

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)
IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

Description Format (URDF). To simplify the description task,

Extensible Markup Language (XML) is used. Inertial mea-
surement is necessary for URDF as it provides data on mass
distribution and inertia properties for each robot component.
This information is required for dynamic analysis, control
design, simulation, collision detection, motion planning, and
system identification. The following equations are used to
correctly calculate the inertial measurement along with the
parameters defined in Table. II

C. Robot Description

TABLE II
ROBOT D ESCRIPTION PARAMETERS ALONG WITH THE VALUES

Parameters Values
Mass of Robot (m) 4.55 kg
Effective Height of Robot (h) 25.6 cm
The effective width of the robot (w 24.5 cm
Wheel Radius (r) 3.5 cm
Wheel Separation (b) 29.5 cm
Acceleration due to gravity (g) 9.8 m/s2
Resistance of Motor (r) 1.2 ohm
Operating Voltage of Robot (Vopt ) 12 V
LIDAR Range 0.15 m - 12 m

These equations are applied in the macros to calculate the

inertia properties for the respective shapes in your robot’s
URDF file. Fig. 4 shows the overall operation of the au-
tonomous robot system as shown in the flowchart. The process
begins with hardware initialization, where the system checks
for readiness by initializing all necessary interfaces. If the
system is not ready, it goes back to the initialization step;
otherwise, it proceeds to sensor data collection. In this stage,
data from the RPLIDAR sensor is gathered and processed.
Once the system is ready, it performs occupancy grid mapping
to create a map of the environment using the RPLIDAR
data. Subsequently, a path planning algorithm is executed Fig. 4. Diagram illustrating the process flow for the robot navigation system.
to determine the optimal path for navigation. If a path is
available, the system controls the motors of the robot using system continuously checks if the goal has been reached. If
an Arduino micro-controller. The system continuously checks the goal is not reached, it loops back to re-plan the path if
if the goal has been reached. If the goal is not reached, it loops necessary. If the goal is reached, the process terminates.
back to re-plan the path if necessary. If the goal is reached,
the process terminates. VII. S YSTEM A RCHITECTURE
Fig. 5 integrates various components to enhance the robot’s
VI. D ESCRIPTION OF ROBOT O PERATION usability and performance. The RPLIDAR A1 captures point
Fig. 4 shows the overall operation of the autonomous robot cloud data, which is processed by the RPLIDAR NODE and
system as shown in the flowchart. The process begins with the 360◦ laser-scanned data is published to the scan topic.
hardware initialization, where the system checks for readiness Once the mapping node receives this data, it triggers a series
by initializing all necessary interfaces. If the system is not of processing tasks on the PC.
ready, it goes back to the initialization step; otherwise, it ROS2 (Robot Operating System) plays a pivotal
proceeds to sensor data collection. In this stage, data from the role in this architecture, particularly through the
RPLIDAR sensor is gathered and processed. Once the system joint_state_publisher_custom, which shares
is ready, it performs occupancy grid mapping to create a map joint information via the /joint_states topic. This
of the environment using the RPLIDAR data. Subsequently, a allows the textttrobot state publisher to convey the data
path planning algorithm is executed to determine the optimal to RViz, where the URDF model, along with the proper
path for navigation. If a path is available, the system controls information from the robot_description, is loaded
the motors of the robot using an Arduino micro-controller. The in the simulation using the my_robot_bringup node.

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 583

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)
IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

Fig. 6. Generation of indoor Static Environment Map.

The joint_state_publisher_custom node provides

the state of wheels, velocity, and position to properly track
the robot’s movement. This alignment with the actual robot
environment is achieved through inertia calculations, ensuring
a realistic and real-time simulation.
An efficient division of work between the Raspberry Pi
and a PC significantly reduces the computing load on the
Raspberry Pi, thereby enhancing the robot’s performance and
efficiency. The Nav2 stack excels in navigation management,
ensuring smooth operations by integrating custom code and
meticulously fine-tuned configuration files. The use of Behav-
ior Trees (BTs) within Nav2 further ensures efficient operation
[12].
Data exchange between the Arduino Mega 2560 and the
Raspberry Pi is handled via serial communication, completing
the robust and efficient design of the 4WDDMR’s structural
and operating framework.

VIII. R ESULT A NALYSIS

A. Mapping

Mapping involves exploring the surrounding area, a crucial

aspect of autonomous navigation. In the paper, two mapping
approaches: one using the SLAM Toolbox and the other with
Google’s Cartographer are evaluated. Cartographer proved
more effective due to its superior optimization and user-
friendliness. As shown in Fig. 6, through RViz visualization,
the mapping technique accurately captured the environment,
distinguishing obstacles (depicted in black) and accessible
Fig. 5. System Architecture consisting of Software and Hardware Components areas (depicted in white). As the robot progressed and LIDAR
for Mapping, Planning, and Control Integrated with Embedded for Sensing mapping continued, the accuracy of the final map improved.
and Actuation.
B. Localization
After mapping, the next step is to localize the robot within
the environment. Utilizing comprehensive environmental data,
particularly odometry information from encoder motors, en-
ables precise localization. As shown in Fig. 7, RViz visualiza-

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 584

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the Fourth International Conference on Sentiment Analysis and Deep Learning (ICSADL-2025)
IEEE Xplore Part Number: CFP25UU5-ART; ISBN: 979-8-3315-2392-3

increases the capability for fine navigation and helps in the

better understanding of operational dynamics.
ACKNOWLEDGMENT
We extend our sincere gratitude to the Department of
Electronics and Computer Engineering, the Department of
Electrical Engineering, and the Robotics Club of Tribhuvan
University, Institute of Engineering, Eastern Regional Campus,
Dharan-08, Nepal, for providing us with the essential resources
and facilities that were crucial in completing this research. We
also wish to express our deepest thanks to Asst. Prof. Manoj
Kumar Guragai of the Electronics and Computer Engineering
Department for his invaluable guidance, unwavering support,
and continuous encouragement throughout the research pro-
cess.
R EFERENCES
[1] M. Quigley, K. Conley, B. Gerkey, J. Faust, T. Foote, J. Leibs, R.
Wheeler, A. Y. Ng et al., “Ros: an open-source robot operating sys-
Fig. 7. Localization testing in the dynamic environment. tem,” in ICRA workshop on open source software, vol. 3, no. 3.2. Kobe,
Japan, 2009, p. 5.
[2] S. Macenski, A. Soragna, M. Carroll, and Z. Ge, “Impact of ros 2 node
composition in robotic systems,” 2023.
[3] S. Macenski, T. Foote, B. Gerkey, C. Lalancette, and W. Woodall, “Robot
operating system 2: Design, architecture, and uses in the wild,” Science
robotics, vol. 7, no. 66, p. eabm6074, 2022.
[4] W. Hess, D. Kohler, H. Rapp, and D. Andor, “Real-time loop closure
in 2d lidar slam,” in 2016 IEEE International Conference on Robotics
and Automation (ICRA), 2016, pp. 1271–1278.
[5] L. Xu, C. Feng, V. R. Kamat, and C. C. Menassa, “An occupancy grid
mapping enhanced visual slam for real-time locating applications in in-
door gps-denied environments,” Automation in Construction, vol. 104,
pp. 230–245, 2019.
[6] A. Buyval, I. Afanasyev, and E. Magid, “Icmv - comparative analysis
of ros-based monocular slam methods for indoor navigation,” in SPIE
Proceedings, vol. 10341. SPIE, March 2017, pp. 305–310.
[7] A. Charroud, K. El Moutaouakil, V. Palade, and A. Yahyaouy, “En-
Fig. 8. Simultaneous localization and mapping of indoor environment. hanced autoencoder-based lidar localization in self-driving vehicles,”
Applied Soft Computing, vol. 152, p. 111225, 2024.
[8] S. Macenski, F. Mart´ın, R. White, and J. G. Clavero, “The marathon
2: A navigation system,” in 2020 IEEE/RSJ International Conference on
tion illustrates the robot’s current position within the mapped Intelligent Robots and Systems (IROS), 2020, pp. 2718–2725.
environment, enhancing navigation precision. [9] E. Marder-Eppstein, E. Berger, T. Foote, B. Gerkey, and K. Konolige,
“The office marathon: Robust navigation in an indoor office environ-
ment,” in International Conference on Robotics and Automation, 2010.
C. Simultaneous Localization and Mapping (SLAM) [10] P. Gyanani, M. Agarwal, R. Osari, and G. Chandore, “Autonomous mo-
bile vehicle using ros2 and 2d-lidar and slam navigation,” 2024.
The goal of SLAM is to estimate the robot’s pose while si- [11] A. Jha and M. Kumar, “Two wheels differential type odometry for
multaneously mapping the environment. As depicted in Fig. 8, mobile robots,” in Proceedings of 3rd International Conference on
this approach involves dynamically updating the map as the Reliability, Infocom Technologies and Optimization. IEEE, 2014, pp.
1–5.
environment changes, enabling the robot to navigate effec- [12] D. L. A. M. S. Macenski, T. Moore and M. Ferguson, “From the desks
tively. By identifying landmarks and estimating the robot’s of ros maintainers: A survey of modern capable mobile robotics al-
pose relative to them, the SLAM technique ensures accurate gorithms in the robot operating system 2,” Robotics and Autonomous
Systems, 2023.
localization and mapping. [13] C. Cadena, L. Carlone, H. Carrillo, Y. Latif, D. Scaramuzza, J. Neira,
I. Reid, and J. J. Leonard, “Past, present, and future of simultaneous
IX. C ONCLUSION lo- calization and mapping: Toward the robust-perception age,” IEEE
Trans- actions on robotics, vol. 32, no. 6, pp. 1309–1332, 2016.
This clearly summarizes the need for autonomous naviga- [14] R. Mur-Artal, J. M. M. Montiel, and J. D. Tardos, “Orb-slam: a versa-
tion made seamless with the aid of Nav2 to access production- tile and accurate monocular slam system,” IEEE transactions on robotics,
vol. 31, no. 5, pp. 1147–1163, 2015.
grade mapping algorithms suited to its particular use cases [15] R. Siegwart, I. R. Nourbakhsh, and D. Scaramuzza, Introduction to au-
using Arduino Mega. When combined with LIDAR technology tonomous mobile robots. MIT press, 2011.
with Cartographer ROS for environmental mapping, fine nav- [16] P. R. Wurman, R. D’Andrea, and M. Mountz, “Coordinating hundreds
of cooperative, autonomous vehicles in warehouses,” AI magazine, vol.
igation becomes achievable. Such data combined with state- 29, no. 1, pp. 9–9, 2008.
of-the-art mapping algorithms allows real-time visualization [17] T. Hellstrom, Kinematics equations for differential drive and articulated
of the position of the robot and its path within RViz. It further steering. Department of Computing Science, Umea University, 2011.

979-8-3315-2392-3/25/$31.00 ©2025 IEEE 585

Authorized licensed use limited to: R V College of Engineering. Downloaded on April 25,2025 at 06:58:33 UTC from IEEE Xplore. Restrictions apply.