CHAPTER 2

LITERATURE REVIEW

2.1 INTRODUCTION

Traditionally, the floor was cleaned by hand with various handcrafted devices. Initially, it was

cleansed with various reed brushes. Egyptian dwellings were said to be made of sun-dried mud

bricks that were sometimes white-washed, and the floors were stamped dirt. The outdoor

kitchen's floor was also made of ground that had been baked stone hard by the sun. These floors

were easy to keep clean by sweeping unless it rained, which happened only seldom. These

brushes, like most ancient Egyptian tools, did not have long handles, which would have made

them easier to use, and therefore required bending low to wield. Brushes of various designs have

evolved for the convenience of humans. During the reign of the monarchs, carpets of various

designs were used to cover the floor and keep it clean. As time passes, a new scientific era

dawns, and a slew of new floor-cleaning procedures emerge. The reciprocating brush action

actuated by muscular force was the first of these. Depending on the floor structure and the ease

of cleaning workers, the brush design is adjusted from time to time. Since the invention of

electricity, floor cleaners have been developed to clean a dry surface. Different floor cleaning

equipment are being developed in the future to clean the floor with minimal bodily effort. Then

there was the idea of a mobile robot. Mobile robots may roam around their environment and are

not restricted to a single physical area. Industrial robots, on the other hand, are normally

stationary, with a jointed arm (multi-linked manipulator) and gripper assembly (or end effector)

attached to a fixed surface.

device was to clean the floor with less power consumption. The sweeping mechanism of the mop

is controlled by a dc circuit and operated by a timed motor. Water is sprayed on the mop, and the

debris is cleaned off the floor with the wet mop. However, it was discovered that no chemical

solvent or disinfection could be used. Again, just hot air is used for soaking. Again, a worker is

required to move the machine. To resolve these issues, a current study was conducted to enable

the cleaner to roam about any room automatically. The moping process has also been tweaked to

save money. In the current study, the mop rotates continually around an axis perpendicular to the

cleaner's action, which helps to distribute water on the floor backward. A sprinkle mechanism is

used instead of a wet mop to make the floor moist, which is then scrubbed by the mop. A floor

cleaner was used to soak up unclean water from the floor surface while cleaning it side by side.

Mobile robotics is used to move the cleaner automatically. Mobile robots are a prominent focus

of current study, with one or more labs dedicated to the field at almost every major institution.

Mobile robots can be found in a variety of settings, including industrial, military, and security.

Domestic robots are consumer items that include entertainment robots as well as robots that

conduct specific household activities like flooring or gardening. Since then, increasingly

advanced robots have been built for domestic devices to automate operations such as washing

machines and micro woven. Only then did the mobile robots revolution reach residential

applications. The problem with today's automatic floor cleaning equipment is that they're only

utilized in homes for dry and wet cleaning, not for removing infections. As a result, it is only

utilized in private homes and not in hospitals or small public spaces. Automatic floor mops, such

as the hydrobot, are huge, demand a lot of electricity, and are utilized for commercial purposes.

However, we believe that this (our cleaning) will solve everything at once. We will primarily
focus on a clever, compact, and well-designed robot that can be utilized in a variety of sectors,

including healthcare and education (both of which are, of course, small places) as well as in the

home. As a result, it will be used for both terminal cleaning, such as medical facilities, and

indoor floor cleaning. As an added feature to this equipment, we will focus on indoor air

purification in the future.

2.2 PARTS OF THE AUTOMATIC FLOOR CLEANER

ROBOT CAR

Robot is an intelligent device having its own brain fed with computer logic so that it can do the

work according to the algorithm designed. Autonomous movement of vehicle is guided by the

logic controller designed. Robots plays an important role in each every field of life. It is used in

industries, in households and in institutes. The robots are just becoming as intelligent as human

now a days. Mostly an average human uses 2-3 robots per day in his day to day life. Various

robotics parts are:-

Pneumatic devices

Actuators

Sensors

Mechanical control devices like valve

Microcontroller – Controlling unit

Mechanical control devices are used to control the flow or movement of materials or any other

parts present in the device. Actuators are used for controlling a mechanism which ultimately
controls a part of the device. Sensors are the sensing devices which transmit a signal and receives

the signal and accordingly used to accumulate the various environment information which is

ultimately fed to microcontroller for deciding the working of machines. Microcontroller is the

brain of robot where program is written and sensors are connected as input and actuators as

output. The controlling of the robot is governed by various algorithms like fuzzy controller,

machine learning based practices and artificial neural network based algorithms. Depending

upon the environment value received to the controller it eliminates the error and transits from

one state to another. Basically there are two types of controllers ,one is continuous controller and

another is PID based controller .Continuous controller is more direct and less effective while PID

controller is more advanced and varies according to the current state and gives efficient result.

SENSORS

Sensor plays a very important role in every type of robot. Various types of sensors are present. A

sensor is a transducer whose intention is to sense (that is, to distinguish) some normal for its

environs. It identifies occasions or changes in amounts and gives a relating yield, by and large as

an electrical or optical sign; for instance, a thermocouple changes over temperature to a yield

voltage. In any case, a mercury-in-glass thermometer is additionally a sensor; it changes over the

deliberate temperature into development and construction of a fluid which can be perused on an

adjusted glass tube. Sensors are utilized as a part of ordinary questions, for example, touch-

touchy lift catches (material sensor) and lights which diminish or light up by touching the base,

other than endless utilizations of which the vast majority are never mindful. With advances in

micro machinery and simple to-utilize microcontroller stages, the employments of sensors have

extended past the more customary fields of temperature, weight or stream measurement, for

instance into MARG sensors. Besides, simple sensors, for example, potentiometers and power
detecting resistors are still generally utilized. Applications incorporate assembling and apparatus,

planes and aviation, autos, solution and apply autonomy. A sensor's affectability shows how

much the sensor's yield changes when the data amount being measured changes. Sensor is

mainly consists of transducers, operational amplifiers. Transducer converts any physical signal to

electrical signal. Basically the electrical signal present in very small range. So we have to use

amplifiers which provides very high gain and increases the signal to volt range. Among various

types of sensor major sensors are ultrasonic distance measurement sensor, sound sensor, pressure

sensor, touch sensor, temperature sensor, and field sensor.

2.2 BRIEF REVIEW OF AUTOMATIC FLOOR CLEANER

The market for robot housemaids appears to be taking off after years of stuttering. The first

evidence of a nascent commercial robot sector gaining hold are new kinds of robot floor

cleaners. However, the robot floor cleaners now on the market are not intended to completely

replace the vacuum cleaner. The cleaners are described as "maintenance" or "continuous"

cleaners by the makers (Kahney, 2003). There are several different varieties of robotic floor

cleaners on the market today, with prices ranging from $50 to $1,800. Because these robots are

often low-slung and tiny, they can clean beneath furniture that ordinary upright floor cleaners

can't reach. Electrolux and Karcher in Europe, and iRobot in the United States, are the three

leading robot floor cleaner firms. Other cleaning producers, such as Hitachi of Japan and Dyson

of the United Kingdom, as well as Friendly Robotics, FloorBotics, Hanool Robotics, Samsung,

and Lentek, are working on the creation of this type of robot. In the United States, iRobot's

Roomba is the most popular robotic floor, with variants ranging from the basic Roomba Red

($150) to the very high-tech Roomba Scheduler ($350). It's a round-shaped floor equipment that

bounces about a room like a pool cleaner, covering all - or nearly all - of the floor (Kahney,
2003). A spinning side brush cleans along walls, two counter-rotating brushes capture large

debris, and a dust-collecting floor are all part of the cleaning system. Roomba returns to the

Home Base to recharge at the end of a cleaning cycle or when the battery is low. When dirt lands

on one of the two sensors immediately above the brushes, it directs the brush's focus to the filthy

region, cleaning it thoroughly. Roomba is also protected from falling down ledges and stairs via

infrared sensors. Virtual barriers provide an infrared signal that Roomba will not pass, allowing

it to clean only where you want it to (Irobot, 2006).

Electrolux's Trilobite is more complicated and costly. As it cleans, the $1800 Trilobite builds an

internal map of the room and automatically recharges when its power reserves are depleted. The

sort of cleaning program you want (regular, quick, or spot) can be selected, and there is a start

button for instant cleaning, or cleaning times can be specified in advance. To recognize and

avoid obstacles, the Trilobite emits harmless ultrasonic waves. It starts by edging along the walls

to map the room's size while cleaning the edges at the same time. After that, it cleans the entire

room. The machine is kept from being stuck by a suspension device, and a particular sensor

keeps it from falling down the stairs (Electrolux, 2006). The RoboCleaner from Kärcher

(Karcher, 2006) costs around 1100€. It takes a "random walk" around the house, using its touch-

sensitive bumpers to detect walls and obstructions. It is low enough to fit under most furniture

and avoids stairwells. It is kept from being stuck by a pair of "rubber ears" on the top. The

RoboCleaner keeps an eye on the incoming dirt stream and focuses on particularly filthy areas.

When the dust bag in the recharging station fills up, the owner merely has to empty it. Dyson, a

British business, is working on a robot cleaner (DC06) and has returned to the R&D phase until

they can create an affordable, self-cleaning floor cleaner. Hitachi, a Japanese electronics

company, is developing a comparable gadget that can also serve as a home security guard. The
robot cleaner may clean itself or be guided by a computer or a smartphone. A built-in camera

allows the owner to keep an eye on his home when he is abroad using the Internet. It has a

charging station where it can recharge its batteries and deposit its dust load. As it walks about the

house, the robot develops a map of the layout. It remembers how the furniture is arranged and

which rooms have been cleaned and which have not. It has a 2-inch retractable hose for cleaning

in tight spaces. Sensors (light, heat, and all-around bumpers) are embedded throughout the

machine to detect hazards and keep it from becoming trapped in corners.

(Manya et al, 2017) An automatic floor cleaner was devised and built. The concept is used to

clean surfaces automatically in both household and industrial settings. It sucks up dust by

moving all over the surface (floor or any other area) when it passes over it when it is turned on.

The controller is used to regulate the motors and suction unit, as well as a few sensors to avoid

obstacles. This has the potential to improve humanity's way of life.

Fig 2.1 Arduino based floor cleaner by (Manya et al, 2017)

robot. In the same breath, home automation has piqued the curiosity of a variety of users.

Currently, industrially made high-priced floor cleaning robots are out of reach for the majority of

third-world households. We have introduced an initiative to construct a cost-effective floor

cleaner robot in this paper. We designed and implemented a cleaning robot using a vacuum

cleaning technology. The paper explains how an operational floor cleaning robot prototype was

created, as well as the indoor mapping and positioning system methodologies. As the robot's

processor, a Raspberry Pi and an Arduino Mega were used in tandem. This robot is capable of

mapping and cleaning a specific region of a building floor on its own. The addition of a GPS

module to the Raspberry Pi is a novel contribution of this work, as it allows the robot to be aware

of its location and move in the correct direction. Only garbage disposal and maintenance require

human assistance. The prototype's future goal is to include a reliable autonomous garbage

disposal feature.

Fig 2.2 Solid works view of an Automatic floor cleaner by (H.A.Shakhawat et al, 2019)
(P.S. Adithya et al, 2018) designed and developed an automatic cleaning and mopping robot.

The cleaning robot assists in cleaning and mopping the floor. This is accomplished by simply

pressing a button, and the robot does the rest. This also reduces the amount of labor used in

factories to clean the floors. As previously stated, this motivated the design and development of

an automatic cleaning and mopping robot that performs all cleaning and mopping tasks with the

simple press of a button. With the help of a mobile Bluetooth device, this robot may be operated

manually. The project's major goal is to make it inexpensive and appropriate for Indian users and

factories. The construction of the robot begins with the creation of a simple and effective robot

chassis, which is a critical component because it must support the robot's entire weight. The

electronics section determines the type of motor and its specifications, as well as the sensors to

be utilized, the microcontroller, motor drivers, wheels, and other electronic components to be

used on the robot. After that, the components will be assembled, and the gadget will be tested

and calibrated. The major goal of the robot is to create a robot that is capable of effective dust

cleaning and mopping of a particular room's floor. Its goal is to make robots affordable and

practical for people of all socioeconomic classes. The robot's target operation time is one hour.

The designed robot will be beneficial in both domestic and industrial settings. This eliminates the

need for physical labor to keep the workstation and home clean. In addition, the device will clean

the room with a single button press.

(Leire M et al, 2019) designed a prototype robot vacuum cleaner. It details, in particular, the

hardware and software components employed, as well as the fuzzy logic-based navigation

algorithm. In addition, an existing virtual prototype of the robot and its surrounding environment

was upgraded in order to fine-tune and test the real controller of the autonomous robot using SIL

(Software-in-the-Loop) simulations.

(Swarup R et al, 2019) designed and fabricated an automatic floor cleaner. He explained that an

automatic floor cleaner is a device that uses a highly stabilized and swiftly functionalized

electronic and mechanical control system to clean the floor. The current project's goal is to

employ an autonomous floor cleaner for large floors in homes and offices. Continuous relative

motion between a scrubber and the floor surface is used to achieve the cleaning goal. During the

cleaning and moving of the vehicle, a propulsion mechanism such as driven wheels and guide

wheels is used for dry tracks on the floor surface to be cleaned, a vacuum pump is used to
suction water, and a scrubber directs water towards the back end. For advance sweeping of a

debris-laden floor surface, a sweeper mechanism placed on the body propelled by the propulsion

mechanism and operated with such a control system is preferred. The motion of the device is

governed by a PID controller, which accepts input from the sensor circuit and transmits it back to

the microcontroller, resulting in a synchronized simulation of the wheel. In the future, the new

robotic floor cleaner will save a lot of money on manpower. The main benefit of this product is

that it will be cost effective and will not require human control. Once turned on, it will clean the

entire room without omitting any surfaces.

(Akanksha V et al, 2013) designed and fabricated a arduino based dry and wet automatic floor

cleaner. The invention of an automatic floor cleaner is the subject of this study. Robotics is

receiving a lot of attention these days as a way to reduce human labor. Our goal is to build a

completely automated floor cleaner that can clean both dry and wet floors while also sterilizing

them with UV light. Cleaners with only one or two functions currently dominate the market.

We're utilizing Arduino because of its low cost and ease of use. The cleaner will be a step toward

making living more comfortable by addressing issues with existing floor cleaning methods.

2.3 Body

To travel automatically, a robot has to know a few details about its coordinates and how to reach

certain destinations and/or way points along the road. A variety of sensors are used to determine

the current location, including motor encoders, eyesight, reacts to warnings, works lifts, and

notifies the war room when an incidence occurs. The SpeciMinder and Tug conveyance robots

for healing center labs are two more independently directed robots, however the latter has people

trained and ready to drive the robot remotely if it loses its autonomy. During the IndaPlant
Project in 2013, craftswoman Elizabeth Demaray and engineer Dr. Qingze developed automation

controlled by plants, as well as a demonstration of trans-species giving. They were able to create

a hybrid substance that was half robot and part plant. Stereopsis, lasers, and global positioning

systems are all examples of stereopsis. Triangulation is commonly used in position-determining

systems. Relative positioning is used to precisely determine the co-ordinates of the machine

component's location and orientation. By gaining access to data that allows it to plot a route to its

next destination or waypoint. It also has access to sensor data that is both time and area

recognized. It can, for example, determine the precise timing and location of radiation levels

when they exceed a safe limit. These robots are frequently part of a remote venture system,

which is linked to the building's various detection and control systems. The PatrolBot security

robot, for example.

2.4 Indoor Navigation

For a robot to be studied indoors, it must consider its surroundings and how it will move in the

not-too-distant future. In the 1970s, wire-direction was used, and in the mid-2000s, reference

point-based triangulation was used. Current commercial robots independently explore in order to

find distinguishing features. Pyxus' HelpMate healing center robot and the Cyber Motion watch

robot, both designed by mechanical autonomy pioneers in the 1980s, were the main business

robots to do this. To investigate structures, these robots first used physically created CAD floor

plans, sonar detection, and divider after variety. MobileRobots' PatrolBot and self-sufficient

wheelchair, introduced in 2004, are examples of cutting-edge technology that can create their

own laser-based maps of a facility and explore open areas and routes. If something gets in the

way, their control architecture adjusts on the go. Initially, the self-sufficient route was chosen

because planar sensors, such as laser extent discoverers, can only perceive at one level. As a
result, robots rely on input from a single sensor and plan their development accordingly. The

most advanced frameworks (such as those developed by IROBOT and SCOOBA) now

incorporate data from a variety of sensors. Smoothness and versatility, for example, can be based

on various sensors in various areas, depending on which provides the most reliable information

at the time, and can self-define the structure. Indoor route robots are not designed to travel

through uncomfortable environments such as stairwells, but they do so on much smoother

surfaces like as floors and glassware. It will be significantly more difficult to complete the task

on rough surfaces. As these indoor systems advance, flooring robots with vapor degreasing will

gain the ability to clean a specific client-defined room or an entire floor while also removing

strain. Security robots have been enhanced so that they may be used securely and reliably at the

entry and exit. These advancements also come with assurances: robots' interior maps typically

allow for the designation of "taboo locations" to prevent robots from self-rulingly entering

particular districts. A flow analysis was conducted in order to design and develop floor cleaning

equipment that can clean every single part of a building organically using an arbitrary calculation

that takes into account the framework guide look.

2.5 Modeling the spiral path

The spiral path is the first step in creating a coverage path. As a result, the robot is at the origin

of the frame at the start of the spiral journey. An imaginary X and Y axis are drawn, which meet

at the robot's beginning position. The spiral can be either clockwise, increasing radius while

decreasing theta, or anti-clockwise, increasing radius while raising theta. Following is a straight

path:- A way that is adjusted to the virtual divider limit is known as a virtual divider

(significance the secured locale) route. A quantic Bezier bend is used to demonstrate this virtual

divider method. The four stages of this way era technique are as follows: inspecting divider
focuses testing way focuses sorting way The molecular swarm streamlining calculation is used to

improve the way. However, at first, the entire path is not formed, but rather each step of the

journey is set in stone. In travel, we precisely maintain seven bent segments. The inspected

divider and way focuses are stored in separate stacks here. We save memory by removing

unnecessary focuses that were used to create previous bend segments from the stack, as well as

from the web. As shown in the accompanying diagram, the secured cells on the outskirts of

secured and undiscovered districts are tested in the first procedure. The inspected divider point is

defined by the focus location of each investigated cell. The guarded district is located on the

right half of the moving path of the example point, while the undiscovered region is located on

the left. The main point to check is the forced opposite separation change from coarse to fine

(CFCIDT). When a specific point is examined twice, the importance circle is completed, or when

each phone around the last tested location converts into a secured or deterrent kind during the

scope, the inspection is completed. When a robot conducts a divider after discovering a snag in

an area during a cleaning activity, it will execute a U-turn or side movements characterized by

connecting two roundabout bends. The surplus floor leeway can be reduced using these two

types of systems. However, the robot's side-movement will be used in restricted areas for

unimportant escape. Nature will be directed to scheduling the cleaning after it has been set up.

The arrangement will be done based on the robot's range and the sweep of turns it will take.

Between these two spans, there must be an unmistakable means to partition. Movement

arrangement (also known as the "route issue" or "piano mover's issue") is a term used in

mechanical autonomy to describe the process of breaking down a desired development project

into discrete movements that meet development constraints and perhaps advance some aspect of

the development. Consider the scenario of a versatile robot researching an inaccessible waypoint
inside a building. It should do this task while avoiding dividers and avoiding crashing down

stairs. A movement planning computation would use a representation of these tasks as input and

generate velocity and turning commands for the robot's wheels. Movement arranging

calculations may be applied to robots with more joints (e.g., modern controllers), more

unpredictable errands (e.g., item control), a variety of needs (e.g., a car that can only drive

ahead), and vulnerability (e.g. flawed models of nature or robot).

In addition to applications in CAD programming, such as invigorating advanced characters,

feature diversion computerized reasoning, building outline, mechanical surgery, and the

investigation of natural particles, movement arranging has a few apply autonomy applications,

such as self-rule, robotization, and robot plan, as well as applications in various fields, such as

invigorating advanced characters, feature diversion computerized reasoning, building outline,

mechanical surgery, and the investigation of natural particles. Diverse locations are shaped in

each district of the residence where the above-mentioned computation will be directed. We can

create a self-contained number of components for a division with a large number of deterrents.

However, given a long passage, there will be many portions with the longest length L that will

cover the most of it. This is referred to as a sectorial structure. Every arrangement is expected to

be associated to a network point in lattice construct approaches, which overlay a matrix with

respect to setup space. The robot is allowed to go to adjacent lattice focuses at each matrix point

if the length of the line connecting them is completely contained inside Cfree (this is tried with

crash location). This discretizes the sequence of events, and inquiry computations are used to

find a path from the start to the finish.

These techniques necessitate establishing a framework. With coarser frameworks, the hunt is

faster, but the calculation will miss paths through tight Cfree partitions. Furthermore, in the
design space measurement, the number of foci on the lattice grows exponentially, making them

unsuitable for high-dimensional problems.

Traditional network-based approaches generate paths whose heading changes are compelled to

be products of a fixed starting point, resulting in faulty paths on a regular basis. Any-point way

arranging techniques find shorter paths by proliferating data along network edges (to pursue

quickly) while avoiding lattice edges (to discover short ways).

When the robot's learning about the setup space changes or the design space itself changes

during way following, network-based approaches typically need to pursue over and over.

Incremental heuristic hunt algorithms re-arrange quickly by using previous similar way

arranging issues to speed up their search for the current one. These methodologies are like

framework based inquiry methodologies aside from that they produce a clearing covering

completely the design space rather than a matrix. The clearing is disintegrated into two sub-

paving X-,X+ made with boxes such that X- ⊂ Cfree ⊂ X+. Portraying Cfree adds up to take

care of a set reversal issue. Interim examination could along these lines be utilized when Cfree

can't be portrayed by straight imbalances so as to have an ensured nook.

After that, the robot is free to travel around in X- but not outside of X+. A neighbor chart is

created for both subpavings, and paths can be found using Dijkstra computations. When a

method is possible in X-, it is also possible in Cfree. When there is no method in X+ to get from

one starting configuration to the goal, we can be confident that there is no way in Cfree. In terms

of the lattice-based technique, the interim methodology is ineffective for high-dimensional

problems since the number of boxes to be generated grows exponentially as the measurement of

arrangement grows.