IOT BASED SMART FLOOR CLEANING ROBOT

FAISAL SHAHKAR

NAEEM SARWAR

AQSA KANWAL

Supervised By
Engr. SAIMA ZAHEER
Submitted for the partial fulfillment of BE Electrical Engineering degree to
the Faculty of Engineering & CS

NATIONAL UNIVERSITY OF MODERN

LANGUAGESISLAMABAD
DECEMBER, 2024
 ABSTRACT
Maintaining clean and hygienic living and working spaces has become increasingly
important in our fast-paced modern world. Traditional floor cleaning methods, such as
manual mopping or semi-automated machines, often require substantial human effort,
time, and supervision. This highlights the need for an intelligent, fully autonomous
solution that can navigate and clean indoor environments efficiently and effectively. The
proposed floor cleaning robot is designed to address these challenges. Equipped with
advanced sensors, a microcontroller, and AI-based algorithms, the robot is capable of
autonomously navigating complex spaces, detecting dirt and debris, and performing
cleaning tasks without human intervention. With its user-friendly design and energy-
efficient system, this robot is intended to provide a cost-effective, practical solution for
homes and businesses. The project leverages technologies like ultrasonic sensors, motor
controllers, and computer vision to enable obstacle avoidance, environmental mapping,
and intelligent cleaning strategies. By automating the cleaning process, this system aims
to improve cleanliness, reduce the physical burden of manual cleaning, and contribute to
better hygiene standards.
 CERTIFICATE
-----------

Dated:

----

Final Approval
It is certified that project report titled ‘IOT Based Smart Floor Cleaning Robot’
submitted by Faisal Shahkar, Naeem Sarwar and Aqsa Kanwal for the partial fulfillment
of the requirement of

“Bachelor’s Degree in Electrical Engineering” is approved.

COMMITTEE

Dr. Omar Hayat

Signature:

Dr. Farhan Sohail

HoD Electrical Engineering Signature:

Engr. Saima Zaheer

Supervisor Signature:

Dr. Sheraz Alam Khan

Signature:
Dr. Rizwan Amirzada
Signature:
Dr. Madah Ul Mustafa
Signature:

1
DECLERATION
 Table of Contents
CHAPTER 1: INTRODUCTION..........................................................................................................

1.1 Introduction......................................................................................................................................

1.2 Problem Statement...........................................................................................................................

1.3 Aims and Objectives........................................................................................................................

1.4 Proposed System..............................................................................................................................

1.5 Development Methodology.............................................................................................................

1.6 Resource Requirement...................................................................................................................

1.7 Social Benefits...............................................................................................................................

1.8 Report Layout................................................................................................................................

CHAPTER 2: BACKGROUND LITERATURE REVIEW..............................................................

2.1 Background....................................................................................................................................

2.2 Literature Review..........................................................................................................................

2.3 Summary........................................................................................................................................

CHAPTER 3: SYSTEM MODEL/PROPOSED SYSTEM................................................................

3.1 SYSTEM MODEL........................................................................................................................

3.2 System Methodology.....................................................................................................................

3.2.1 Obstacle Detection..................................................................................................................

3.2.2 Remote Control and Monitoring.............................................................................................

3.2.3 Fire Detection.........................................................................................................................

3.3 List of Components........................................................................................................................

REFERENCES......................................................................................................................................

2
 LIST OF FIGURES
Figure Caption Page
3.1 System Model...............................................................................................................................

3
 LIST OF TABLES

Table Caption Page

3.1 List of Components.......................................................................................................................

4
 CHAPTER 1
INTRODUCTION

5
1.1 Introduction
The field of robotics has been growing rapidly, driven by advancements in artificial
intelligence (AI), sensor technology, and microcontroller systems. Among the various
applications of robotics, autonomous floor cleaning has become a significant area of
interest due to its potential to save time and labor, while maintaining high cleanliness
standards. Cleaning robots have gained popularity for their ability to navigate complex
indoor environments, detect dirt, and perform their tasks without the need for human
oversight. With the rising demand for more efficient and user-friendly cleaning solutions,
the development of floor cleaning robots that can cater to both households and
commercial spaces has become increasingly important [1].

The technology behind autonomous cleaning robots involves the integration of hardware
and software systems that enable them to move, map their environment, detect dirt or
debris, and clean effectively. Ultrasonic and infrared sensors are typically used for
navigation and obstacle avoidance, while motor systems facilitate the robot’s movement
across the floor. To further enhance the robot's functionality, some models may
incorporate computer vision or LIDAR sensors for more precise mapping and decision-
making capabilities. The cleaning mechanism often includes brushes, rollers, and suction
systems to ensure that floors are effectively cleaned, whether they are hardwood, tile, or
carpet.

Automated cleaning solutions have shown benefits not only in terms of convenience but
also in terms of cost and efficiency. Households with busy schedules and commercial
spaces that require regular cleaning can benefit significantly from the use of autonomous
robots. These robots minimize the need for human labor, reduce the risk of human error,
and improve the consistency of the cleaning process. However, despite the advances,
existing robots still face challenges such as navigating irregular floor plans, detecting
different levels of dirt, and operating for extended periods without needing a recharge.
These challenges present an opportunity to design a more advanced system that addresses
these limitations and provides a reliable, cost-effective solution.

6
The motivation behind this project is to design and develop a floor cleaning robot that can
navigate a variety of indoor spaces, detect varying levels of dirt, and clean autonomously
without human supervision. The robot should be efficient, energy-conscious, and capable
of performing cleaning tasks across different floor types. By leveraging a combination of
sensors, motors, and AI algorithms, the aim is to create a system that provides superior
navigation, intelligent path planning, and an effective cleaning process. This project
represents a step toward making fully automated floor cleaning accessible and practical
for everyday use in households and commercial settings [2].

1.2 Problem Statement

The current reliance on manual labor and traditional cleaning methods for floor
maintenance presents several challenges, including inefficiency, safety risks,
environmental impact, and inconvenience. Manual cleaning is time-consuming, labor-
intensive, and can pose safety risks such as slips, trips, and falls. Additionally, traditional
cleaning methods often involve the use of harsh chemicals that can contribute to indoor
air

pollution and environmental contamination. Furthermore, the repetitive nature of floor

cleaning tasks can be tedious and affect overall well-being.

To address these challenges, there is a pressing need for an innovative and efficient
solution that can automate floor cleaning tasks, enhance safety, improve environmental
sustainability, and provide convenience and flexibility. An autonomous floor cleaning
robot equipped with advanced features such as obstacle avoidance, fire detection, and
remote control can offer a viable solution to these challenges. By reducing the burden of
manual labor, minimizing safety risks, promoting eco-friendly cleaning practices, and
providing a user-friendly interface, such a robot can significantly improve the efficiency
and convenience of floor maintenance. The manual cleaning methods is inefficient and
labor-intensive. There is a critical need for an innovative solution that automates floor
cleaning tasks.

7
 1.3 Aims and Objectives
The aim of the project is to develop a versatile and intelligent multi-purpose IoT based
robot (MAROBOT) with mopping and vacuuming capabilities for enhanced safety,
efficiency, and convenience in cleaning tasks.

Objectives: The objectives of this project is

• To implement an efficient algorithm for autonomous floor cleaning, including navigation

and obstacle avoidance.

• Integrate advanced fire detection sensors and implement a reliable alert system and utilize
appropriate sensors and algorithms for accurate obstacle detection and avoidance.

• Create a user-friendly interface for remote control and monitoring.

1.4 Proposed System

The proposed system is an advanced, portable floor cleaning robot designed to automate
the process of cleaning indoor surfaces with high efficiency and adaptability. The robot
will be equipped with a variety of sensors and intelligent algorithms for autonomous
navigation, dirt detection, and comprehensive floor cleaning. The system will include a
microcontroller (such as Raspberry Pi or Arduino) at its core to process data and control
the operation, along with an array of essential hardware components that support its
functionality.

The project specifications include a combination of sensors, motor controllers, and a

cleaning mechanism integrated into a compact design. The robot will be equipped with
ultrasonic and infrared sensors for real-time mapping of its environment, enabling it to
detect obstacles, create a navigable path, and avoid collisions. This mapping and path-
planning will be managed by specialized algorithms that use sensor data to generate a
floor plan and decide the most efficient cleaning route. The system will incorporate an
good and

autonomous path-following algorithm to ensure thorough coverage of the floor area

without human intervention.

8
A powerful microcontroller, such as a Raspberry Pi, will be used to process the data from
the sensors and manage the robot's movement and operations. The software will be
programmed to include object recognition and obstacle-avoidance algorithms, ensuring
that the robot can effectively navigate around furniture and other objects on the floor. The
system will feature an integrated cleaning mechanism, including rotating brushes and a
suction system for efficient dirt collection [3].

The robot will also come with an automated docking system for recharging its power
source. A rechargeable battery pack will provide the necessary energy for operation,
enabling the robot to work continuously for an extended period before returning to its
charging station. A mobile app or remote interface will be included for user control and
monitoring, allowing users to start, stop, or schedule cleaning tasks at their convenience.

The entire system will be designed with portability and user-friendliness in mind,
allowing it to be adapted for various home and office environments. The integration of
these technologies aims to make the cleaning process faster, more efficient, and
accessible, contributing to improved living and working conditions for users by reducing
the physical effort needed for routine cleaning. The system's compact size and modular
design will ensure that it can fit into tight spaces, providing a flexible cleaning solution
for diverse household or commercial settings.

This floor cleaning robot will be a practical and cost-effective solution, particularly
suitable for users who want to enhance their cleaning practices without investing in high-
priced, proprietary robotic models. The proposed system is poised to make a significant
impact by promoting a cleaner and healthier living environment, supporting better
hygiene practices while maintaining convenience for the user.

1.5 Development Methodology

The methodology for this project begins with assembling the basic hardware components
necessary for building the floor cleaning robot and ensuring its initial functionality. The
9
primary focus at this stage has been connecting the tires to the chassis frame, which forms
the main structure of the robot. This configuration allows for basic movement and serves
as the foundational base for integrating additional components.

The robot is equipped with a variety of essential components for obstacle detection,
safety, and cleaning operations. To detect obstacles in real-time and help navigate the
robot around various obstacles, ultrasonic sensors are utilized. These sensors provide data
that enable the robot to measure the distance between itself and nearby objects, allowing it
to adjust its path and avoid collisions effectively.

The control and coordination of the robot's movement are managed by a microcontroller,
which processes the signals from the sensors and sends commands to the motor driver
board. The motor driver board (L293 motor driver) is used to control the motors
connected to the tires, enabling the robot to move in different directions and maintain
stability while navigating through the environment. The motor driver board is essential
for managing the power supplied to the motors, ensuring that they operate smoothly and
efficiently.

Safety and alerting mechanisms have also been incorporated into the robot's design. A
buzzer is integrated into the system to provide auditory alerts, which can be triggered by
specific events, such as an emergency or potential malfunction. Additionally, a fire sensor
is included for detecting the presence of smoke or fire, enhancing the safety of the
environment where the robot operates.

For remote communication and control, an HC-05 Bluetooth module is employed,

allowing the robot to connect with external devices, such as a smartphone or computer.
This connectivity enables users to send commands to the robot and monitor its status
remotely, facilitating ease of use and greater control.

The robot's cleaning capabilities are supported by a vacuum system powered by a suction
motor, which collects dirt, dust, and debris from the floor as the robot moves. The suction

10
 motor is controlled through a 5V single-channel relay module, allowing it to be activated
and deactivated as needed. To ensure a stable power supply, a 7805 voltage regulator is
used to maintain the appropriate voltage levels for the microcontroller and other low-
voltage components. The entire system is powered by a dry battery, which provides the
necessary energy for operation.

This stage of development establishes the core functionality of the floor cleaning robot,
including basic movement, obstacle detection, safety alerts, and cleaning capabilities. The
next steps will involve refining these systems, implementing additional sensors and
algorithms for path planning, and enhancing the robot's ability to clean more effectively
and efficiently.

1.6 Resource Requirement

The resource requirements for this project include both hardware and software
components essential for building and operating the floor cleaning robot. We are using
microcontroller
• to processes data and coordinates the actions of other components. Ultrasonic sensor are
used for obstacle detection and measuring distances to ensure safe navigation and
collision avoidance. We are tyres to provides the movement for the robot, allowing it to
travel and maneuver across the floor. We are sing L293 Motor Driver Board to controls
the motors, ensuring proper functioning and direction of movement. Buzzer and Fire
Sensor will be used as an alert system to provide audio notifications for specific events or
warnings and detects the presence of smoke or fire, contributing to the safety of the
environment where the robot operates. HC-05 Bluetooth Module will enables wireless
communication between the robot and external devices, such as smartphones or
computers, for remote control and status monitoring. 5V Single-Channel Relay Module
and 7805 Voltage Regulator will be used to controls the activation and deactivation of the
vacuum system and other low-voltage components and maintains stable voltage levels for
the microcontroller and other electronic parts to ensure consistent performance. Dry
battery and Suction motor for vacuum will be used to provides the necessary power
supply to run the entire system and powers the vacuum system, allowing the robot to pick
up dirt, dust, and debris from the floor.
11
1.7 Social Benefits
This project has significant social benefits that contribute to public safety and
convenience. The floor cleaning robot equipped with obstacle detection, safety features,
and a vacuum system can help maintain cleaner and safer environments in public spaces.
By autonomously cleaning floors in high-traffic areas, this robot can reduce the risk of
slips and falls, contributing to a safer environment for pedestrians and reducing potential
injuries.

Moreover, this system can be especially beneficial in places like hospitals, shopping
malls, and large commercial facilities where maintaining cleanliness is crucial for the
well-being of visitors and staff. The use of automation in cleaning not only enhances
efficiency but also reduces the manual labor required, which can help minimize work-
related injuries and improve workplace conditions for cleaning staff.

The integration of safety features such as the fire sensor and buzzer also adds an extra
layer of security, alerting users to potential hazards and enabling swift action to prevent
accidents or damage. This can be particularly valuable in public or institutional settings
where early warning systems can mitigate risks and improve overall safety.

Overall, the floor cleaning robot offers a practical solution that supports public health and
safety, making it easier to maintain clean and safe environments with minimal human
intervention. By leveraging technology, this project seeks to improve quality of life and
contribute positively to both commercial and residential spaces.

This project supports Goal 9 by integrating innovative technologies such as obstacle

detection, automation, and real-time safety features to create a robust, efficient floor
cleaning system. The use of microcontrollers, sensors, and motor drivers enables this
system to function autonomously, demonstrating the potential for resilient infrastructure
in cleaning and maintenance. By promoting automation in public and commercial spaces,

12
this project fosters sustainable industrialization and innovative approaches to facility
management, reducing reliance on manual labor and enhancing productivity.

This project contributes to Goal 11 by promoting safety and sustainability in urban

environments. By providing real-time obstacle detection and alert mechanisms, the floor
cleaning robot can operate in high-traffic public areas without posing a risk to pedestrians,
thereby making cities safer and more inclusive. The robot’s vacuum functionality helps
maintain clean and hygienic public spaces, contributing to a healthier urban environment.
Moreover, the implementation of this system reduces the need for intensive manual labor,
which can lead to more sustainable and efficient facility operations, ultimately enhancing
the resilience of human settlements.

1.8 Report Layout

The first chapter of this report includes the introduction, problem statement, aim and
objectives, methodology, resource requirements, social benefits, and alignment with SDG
goals. It sets the stage for the entire project by providing an overview of the problem
being addressed and the goals set to achieve it. The second chapter delves into the
background and history, providing a comprehensive review of the current body of work in
the field, highlighting existing solutions and gaps that this project aims to fill.

The third chapter is dedicated to describing all the components and technologies used in
the project. It offers detailed information about the hardware and software elements
involved, such as the sensors, microcontrollers, and modules that enable the system to
function efficiently. The fourth chapter covers the circuit diagram, detailed descriptions of
the hardware setup, software requirements, and how each component interacts with others
to create a cohesive system.
The fifth chapter focuses on testing and verification, detailing the procedures followed to
ensure that the system meets all expected requirements and functions as intended. It
discusses the results of the tests conducted and addresses any challenges encountered
during the testing phase.

13
Finally, the sixth chapter wraps up the report with a conclusion summarizing the project’s
achievements and its overall impact. It also explores the future scope of the project,
suggesting potential areas for further development and improvements to extend the
system’s capabilities and applications.

14
 CHAPTER 2
BACKGROUND LITERATURE REVIEW

15
2.1 Background
Cleaning robots have evolved significantly over the years, combining advancements in
robotics, artificial intelligence, and automation to provide efficient and convenient
cleaning solutions. This article presents a chronological overview of the development
history of cleaning robot products, highlighting key milestones and innovations along the
way.

The concept of cleaning robots emerged in the 1960s with the introduction of simple
floor-cleaning machines. These early machines used basic sensors to detect obstacles and
navigate across surfaces. However, they lacked intelligence and were limited in their
capabilities.

In the 1990s, the first automated vacuum cleaners entered the market, revolutionizing
household cleaning. The most notable pioneer was the Electrolux Trilobite, launched in
1996, which employed sensors to detect and navigate obstacles, ensuring efficient
cleaning without manual intervention.

The 2000s witnessed significant advancements in navigation and mapping technologies

for cleaning robots. The introduction of simultaneous localization and mapping (SLAM)
techniques enabled robots to create real-time maps of their surroundings, improving their
ability to navigate complex environments. The iRobot Roomba, launched in 2002, utilized
SLAM technology to enhance its cleaning capabilities.

The 2010s marked a major milestone in the development of cleaning robots with the
integration of artificial intelligence (AI) technologies. AI algorithms enabled robots to
learn and adapt to their environment, improving their cleaning performance over time.
Companies like Neato Robotics and iRobot incorporated AI into their cleaning robot
products, enhancing their ability to detect and clean specific areas more efficiently.

The current decade has witnessed the integration of cleaning robots with smart home
ecosystems. With the rise of Internet of Things (IoT) technology, cleaning robots can be

16
controlled remotely via smartphone apps or integrated with voice assistants like Amazon
Alexa or Google Assistant. This connectivity allows users to schedule cleaning sessions,
monitor cleaning progress, and customize settings from anywhere, enhancing convenience
and automation.

In recent years, there has been a growing emphasis on developing cleaning robots that are
environmentally friendly and sustainable. Manufacturers are incorporating energy-
efficient components, using eco-friendly materials, and designing robots for easy
disassembly and recycling, reducing their environmental impact [4]. The desired
functions and the design of an ultimate versatile personal robot is an ongoing debate.
Meanwhile, however, precursors of this yet to evolve species are well on their way to
become commercial products. Cleaning robots for public environments as well as for
private households seem to be able to provide the breakthrough which the designers of
non-industrial robot systems have long awaited.

This survey describes a selection of 30 different cleaning robots, with the first
developments reaching back more than 15 years. With a few exceptions we have focused
on floor cleaning, in particular indoor floor cleaning. We describe a variety of scrubbing
and vacuuming robots which were developed for this task. The described systems range
from heavy, large, and expensive industrial cleaning vehicles to small-size, light-weight,
low-cost household devices. The survey does not include, however, systems for cleaning
facades of buildings, or windows, or production tools.

Although not all of the 30 cleaning robots abovementioned have yet reached the state of
commercial products, their number alone certainly reflects the expectations regarding the
economic value associated with the automation of cleaning tasks. In Europe only the
estimates for the market for cleaning services range up to the order of US$1100 billion
per year. It is therefore not surprising respect to the automation of cleaning tasks using
(semi)autonomous mobile robot systems [5].

17
2.2 Literature Review
Android-controlled voice floor-cleaning robots were developed by researchers. They put
together key hardware and software parts. It used a module that communicated with an
Android smartphone through Bluetooth HC-05 wireless communication. Two DC motors
were attached, each provided with an L293D motor driver IC, receiving instructions either
to move forward or backward. With sponges attached to the end, these added another
advantage to cleaning. A smartphone, using the "BT Voice Control for Arduino" app, was
programmed through the Arduino IDE, with a C language processor of voice commands.
Four 1.5V batteries combined into a 6V power source to maintain consistent functioning.
Testing proved that the robot successfully carried out movements with accuracy, therefore
proving that all components were properly integrated into a working and responsive
cleaning prototype [6].

A rechargeable 12V Lithium Polymer battery has been used along with an Arduino Mega
microcontroller for the system, which provides broad GPIO pins and high memory in designing
and developing the device. The system is comprised of a dual cleaning mechanism such as a
vacuum system for collection of dust and a mopping system for removal of dirt, which further
contains detergent. It used ultrasonic and infrared sensors for obstacle detection for smooth
navigation. The Android mobile application was developed in Android Studio that enabled
Bluetooth-based control within a 10-meter range. Google Firebase gave a secure login and
authentication system. Thorough testing exhibited efficient operation, both autonomous and
manual modes. Even after long durations of time, stable performance was maintained; however,
some minor issues like skidding of wheels happened only after prolonged usage. In the future,
some of the proposed improvements include smart navigation through AI integration, voice
assistants such as Alexa, solar modules for long usage, and Wi-Fi connectivity for real-time
monitoring and data analysis [7].

Using an Arduino Mega microcontroller, that is to say a very broad GPIO pin with large
memory, for the system and a rechargeable 12V Lithium Polymer battery to create and
develop a smart, autonomous cleaning and mopping robot which is equipped with a
double cleaning mechanism in the form of a vacuum system to collect dust and a mopping
system to remove dirt that has been improved by adding detergent. It makes use of
ultrasonic and infrared sensors to detect obstructions for smooth navigation. Android
18
mobile application was developed in Android Studio that enabled Bluetooth-based control
within a 10-meter range. Google Firebase gave a secure login and authentication system.
Proper testing ensured the efficient operation of both autonomous and manual modes.
Stable performance was achieved even for extended periods, but minor problems such as
skidding of wheels came up only after considerable usage. Some future proposals with
regard to its improvements include smart navigation by adding AI, voice assistants like
Alexa, and solar modules to use for longer durations and Wi-Fi for real-time monitoring
and analytics [8].

This implementation of a strong vision system on floor-cleaning robots allows it to

optimize their navigation and the analyzing the surroundings of the floor that,
consequently can reduce consumption in terms of power, water, and chemical products.
This paper proposes a novel pipeline of a vision system that may be integrated into a
floor-cleaning robot. This vision system was built on the YOLOv5 framework and its
purpose is to detect dirty spots on the floor. The two cameras feeding the vision system
are located on the front and back of the floor-cleaning robot. The front camera should
save the energy and resources of the floor-cleaning robot by controlling its speed and the
amount of water and detergent spent in accordance with the dirt detected. The objective of
the rear camera is to serve as an assessment and1assist the navigation node because it will
enable the robot to understand if it has effectively cleaned the floor or needs to return later
to clean again. Both cameras had implemented a self-calibration algorithm that was aimed
at stabilizing image intensity and increasing the robustness of the vision system. A
YOLOv5 model was trained using a wellprepared training data. A new dataset was
acquired from the floor-cleaning robot in an automotive factory. The training dataset used
was hybrid and consisted of the dataset of ACIN, the dataset from the automotive factory,
and the synthetic dataset. Data augmentation was performed in order to expand the dataset
and to balance the classes. Finally, our vision system reached a mean average precision
(mAP) of 0.7 on the testing set. Keywords Computer Vision Object Detection Deep
Learning Floor-cleaning robots [9].

This cleaning and mopping robot uses a Raspberry Pi 3 as its central controller, which
connects with L298N motor drivers that operate four 12V DC motors to move and clean.
19
Li-ion batteries are used to power the robot. It has a cleaning module, which includes
rotating scrubbers, a wiper, and a water pump motor to dispense cleaning liquid.
Controlled by the Blue Dot Android application, the user can easily control the robot in
forward, backward, left, and right directions with the mopping and sweeping functions.
The design has been very harmonious in terms of hardware and software integration; it
uses the Raspberry Pi OS with Python-based control programs. Indoor testing
demonstrated efficient cleaning, confirming the robot's functionality and effectiveness in
achieving its tasks [10].

This COVID-19 pandemic has led to the need for better cleaning solutions than just
dusting. This is with a focus on sanitization and less human labor to avoid infection. This
led to the designing of an autonomous disinfecting robot that can clean and sanitize the
environment effectively with minimal human interference. The robot has an integrated
module of advanced components including Raspberry Pi 4, Arduino Mega 2560, LIDAR
sensors, and a cleaning mechanism using motor- driven sponges and a disinfectant spray
system. It is constructed from lightweight aluminum in a modular design that ensures the
durability and efficiency of the robot. Stress analysis was done with ANSYS to ensure the
load-bearing capacity of critical components like the acrylic plate. The robot uses ROS
and SLAM technology and efficiently navigates, cleans, and avoids obstacles. Testing in a
10'x11' room showed the efficiency of cleaning to be 90%, with improvement suggestions
being made on corner coverage with extended foam rollers. It is a new solution to the
modern cleaning challenge while emphasizing safety and hygiene [11].

The authors developed a selective area cleaning framework for the hTetro reconfigurable
robot, using RGB-D sensors, deep learning algorithms, and evolutionary optimization
techniques. The system detects dirty areas by combining human traffic patterns, tracked
with the Deep SORT algorithm, and stain/trash identification, via SSD MobileNet, to
obtain high accuracy in real time. The detected dirty spots are converted into 3D
coordinates for precise localization. An optimal path is generated by an evolutionary
optimization algorithm that guides the robot efficiently through these areas, minimizing
navigation time and energy consumption. The reconfigurable design of the hTetro robot
enables it to change its shape to clean narrow spaces, thus making it more versatile.
20
Thorough testing in diverse indoor environments showed the framework's ability to work
under different lighting conditions and significantly improve cleaning precision. This
method also surpassed conventional cleaning techniques in terms of accuracy, energy
efficiency, and time savings, which proved its practical value for autonomous cleaning in
real-world scenarios [12].

The article introduces a grid-based offline Coverage Path Planning strategy underlining
robot safety; for this purpose, it adopts the Failure Mode and Effect Analysis technique to
explore risk values in a robot environment. Upon these values, the FMEA strategy guides
the CPP algorithm to avoid areas with high risk values. This approach differs from
traditional CPP methods in that it classifies hazards based on risk levels, effectively
managing the trade-off between maximizing area coverage and ensuring safety. The
proposed method is validated using the hTetro mobile robot, and results demonstrate its
ability to reduce high-risk movements compared to existing methods. The framework can
be further improved through additional experimentation and integration with AI- based
learning for adaptive navigation [13].

2.3 Summary
The smart floor cleaning robot based on IoT is an innovative project that focuses on
designing and developing intelligent autonomous cleaning systems for indoor
environments with navigation, dirt and debris detection, and efficient cleaning activities.
The system integrates such essential hardware components as a microcontroller (Arduino
or Raspberry Pi), ultrasonic sensors, a camera, motors, wheels, a rechargeable battery, a
sturdy chassis, an alert mechanism, advanced fire detection sensors, and a GSM module.
AI-based algorithms will improve its ability to process sensor and camera data in real-
time for obstacle detection, autonomous decision-making, and optimized cleaning paths.
The robot will have features such as fire detection, alerts, and remote monitoring and
control through a user-friendly smartphone application. The methodology involves
hardware assembly, dataset collection, sensor integration, algorithm development, and
system testing. This project is intended to develop efficient algorithms for autonomous
cleaning, incorporate advanced safety features, and offer a remote-control interface to
improve the user experience. The project is in line with the Sustainable Development

21
Goals 9 and 11 since it fosters innovation, infrastructure development, and sustainable
urban living. The social benefits from this project include improving public safety,
reducing manual labor, improving hygiene inside homes, and increasing convenience in
everyday life. It expands on previous work in computer vision, machine learning, and
smart home integration with the goal of making substantial inroads into the advancement
of autonomous cleaning technology to create smarter, safer, and more efficient solutions
for homes.

CHAPTER 3
SYSTEM MODEL/PROPOSED SYSTEM

22
3.1 SYSTEM MODEL
The proposed hardware architecture integrates several components for efficient operation.
A smartphone application serves as the user interface, simplifying control and
monitoring. The microcontroller, which serves as the central processing unit, manages
inputs from the camera to send live video, obstacle detection sensors to direct navigation,
and fire detection to guarantee security. The outputs include motor control for movement
via a motor driver and vacuum and mop capability for cleaning tasks. Power is supplied
to every component, ensuring the seamless integration and operation of the system. This
design creates a comprehensive solution for automated cleaning and monitoring tasks.

Figure 1: System Model

23
3.2 System Methodology
3.2.1 Obstacle Detection
The obstacle detection dataset consists of images and sensor data from various
environments, such as houses and workplaces. Using these images and sensor data, the
robot is trained to recognize obstacles in its path, including walls, furniture, and other
items. The collection includes images taken from several angles and in different lighting
situations to ensure accurate detection. For this purpose, a diverse dataset has been
created via both Internet sources and unique data collection methods. In real time, the
dataset teaches the robot to identify and avoid obstacles throughout the cleaning process.

A dataset comprising multiple floor kinds, including concrete, wood, carpet, and tiles, is
obtained for the purpose of recognizing the type of floor. The robot can recognize the
type the floor and modify its cleaning approach due to the dataset's photos as well as
sensor data. For instance, the robot vacuums carpeted floors and mops wooden flooring.
To provide a wide variety of floor types for the training process, this dataset was
collected from various sources, such as homes, work environments, and public spaces.

The floor-cleaning robot uses ultrasonic, infrared, and camera sensors to identify
impediments. Data from many sensors is received by the microcontroller (such the
Raspberry Pi), which interprets it and uses it to make real-time navigational decisions.
The motor driver controls the robot's movement, ensuring that it avoids obstacles and
occupies the whole cleaning area. The camera records visual information to detect
impediments. After that, the robot adjusts its position to offer the best service.

The robot's cleaning functions are separated into vacuum and mop functions. A dust
sensor determines how much dirt is on the floor and initiates the proper cleaning
procedure. The robot activates the suction motor to gather debris when it detects dirt. The
mop is turned on when wet cleaning is taking place. In order to enhance cleaning
performance and transition between cleaning modes, the microcontroller evaluates sensor
feedback.

24
3.2.2 Remote Control and Monitoring
The floor-cleaning robot can be controlled via the smartphone application. Users may
control the robot's cleaning settings, keep an eye on its condition, and get notifications
while the robot is cleaning. Alerts can also be sent to users, for example, when the robot
identifies an obstruction or finishes cleaning.

The mobile app continuously receives information about the robot's condition, such as its
battery level, cleaning progress, and fault warnings. The smartphone app logs the robot's
data to the cloud and offers real-time monitoring. The cleaning area, any barriers found,
and the amount of time spent cleaning are all included in this report. Data accessibility
for analysis and troubleshooting from a distance is guaranteed via cloud storage.

3.2.3 Fire Detection

To guarantee safety while in use, the robot has a fire detection sensor. If a fire or smoke
is detected, the system instantly notifies the user through the mobile app and initiates an
emergency cleaning system shutdown. This is an essential safety feature, especially in
areas where flammable materials are present.

3.3 List of Components

The components that will be used in this project are listed in Table 3.1

Sr.N Components
o
1 Arduino,
Raspberry Pi
2 Ultrasonic
Sensors
3 Camera
4 Motors and
Wheels
5 Battery
6 Chassis
7 Alert
Mechanism
8 Fire Detection
Sensors

25
 9 GSM Module
Table 1: List of Components

REFERENCE

26
[1] G Tuangzhi Dai and Tiequn Chen "Design on measurement and control system of cleaning robot
based on sensor array detection", IEEE International conference on control automation
Guangzhou, May 30 to June 1, 2007, China.

[2] Youngkak Ma, seungwoo Kim, Dongik Oh and Youngwan Cho, "A study on the development of
home mess- cleanup robot McBot", IEEE/ASME international conference on advanced
mechatronics, July 2-5, 2008, Xian, China.

[3] V.Prabakaran, M.Rajesh Elara, T.Pathmakumar and S. Nansai "HTetro: A Tetris inspired
shapeshifting floor cleaning robot", IEEE International Conference on Robotics and Automation
(ICRA), May 29 to June 3, 2017, Singapore .

[4] "www.reemanrobot.com/news," reemanrobot, Jun 27 2023. [Online]. Available:

https://www.reemanrobot.com/news/development-history-of-cleaning-robot-products-
68788278.html.

[5] A. R. C. S. P. F. I. &. C. Erwin Prassler, "Autonomous Robots/Vol. 9, No. 3A," ACM Digital Library,
[Online]. Available: https://dl.acm.org/doi/10.5555/591546.591832.

[3] M. D. S. W. R. Ruri Ashari Dalimunthe, "Floor Cleaning Robot Control System with Android Based
Voice," IOP Publishing, STMIK Royal Kisaran, Indonesia, 14/02/2021.

[6] A. P. M. &. P. K. Ramkissoon, "A smart autonomous floor cleaner with an Androidbased
controller," Anshu Prakash Murdan, Réduit, Mauritius, 14 June 2023.

[7] A. H. Kassim, "Internet of things-based floor cleaning robot," Indonesian Journal of Electrical
Engineering and Computer Science, Kuala Terengganu, Malaysia, December 2023.

[8] P. F. P. G. A. J. R. N. Daniel Canedo, "publication," Researchgate, April 2022. [Online]. Available:

https://www.researchgate.net/publication/
360184474_An_Innovative_Vision_System_for_FloorCleaning_Robots_Based_on_YOLOv5.

[9] M. V. S. R. R Ananya, "publication," Researchgate, October 2023. [Online]. Available:

https://www.researchgate.net/publication/
386346670_DESIGN_AND_IMPLEMENTATION_OF_A_BLUETO
OTH_CONTROLLED_FLOOR_CLEANING_ROBOT.

[10] K. O. A. A. M. A.B. Baratova1, "THE FEATURE OF THE AUTONOMOUS ROBOT FOR CLEANING
THE," U. Joldasbekov Institute of Mechanics and Engineering, Kazakhstan, Almaty, 2022.

[11] A. L. L. Balakrishnan Ramalingam1, "Optimal selective foor cleaning using deep learning
algorithms and reconfgurable robot hTetro," Scientifc Reports, Nanyang Technological,
University, Singapore 639798, Singapore..

[12] S. M. B. P. S. M. A. V. J. M. Isira D. Wijegunawardana, "FMEA-Based Coverage-Path-Planning

Strategy for Floor-Cleaning Robots," 2023.

27
[13] K. O. A. A. M. A.B. Baratova1, "THE FEATURE OF THE AUTONOMOUS ROBOT FOR CLEANING
THE," U. Joldasbekov Institute of Mechanics and Engineering, Kazakhstan, Almaty, 2022.

28