Aman George Sebastian, 2024, 12:2 International Journal of Science,

ISSN (Online): 2348-4098 Engineering and Technology

ISSN (Print): 2395-4752
An Open Access Journal

Water Surface Cleaning Robot

Aman George Sebastian, Madhav Devnarayan, Bachu James, Abhishek Thomas
1
Dept. of elctronic engineering Saintgits College of Engineering(Autonomous)Kottayam, Kerala
2,3,4
Robotics and Automation Engineering Saintgits College of Engineering(Autonomous) Kottayam, India

Abstract- We all depend on water for many things in our everyday lives. It is a fantastic, necessary source of

life. For many towns and communities, rivers and other bodies of water continue to be a significant supply of
drinking water. The amount of trash in these bodies of water, however, is more than simply a nuisance; it poses

a threat to nature, our lives, and the lives of those we love. Even a single item of litter left on the ground can

add to the buildup of trash in our rivers and creeks. Although the water is initially cleansed before it enters our
houses, if these water bodies are still polluted, the water cannot be sufficiently purified to become unfit for

human consumption. To ensure that water continues to flow from our taps, it is essential to maintain our river
systems and keep them free of pollutants. These serve as the driving forces behind our project. We are working

to create an autonomous water surface cleaning robot that will move through the aforementioned water

bodies in an effort to gather rubbish that floats in them. Our main goal is to keep the water bodies clean
without human supervision. The region will be given to the robot, which will navigate it while floating across

the area and collecting the aforementioned garbage. We also intend to mount a camera on the robot to

provide real-time visuals and detect any waste products present in the water body. A conveyor belt connected
to the boat is to be used to collect the waste materials. The waste materials are conveyed on a conveyor belt to

a collecting area, where they are gathered and kept until the boat docks again

Keywords- Floating waste products; Autonomous; Con- veyor belt;

I. INTRODUCTION robotics technology capable of operating in water

areas. The proposed applied research is expected to
be an alternative solution to prevent disasters,
Waste is an environmental problem that always
especially floods. Robotics technology developed in
arises from year to year and still cannot be resolved
the form of eco-robot with the main task of
entirely. We frequently found garbage from various
collecting waste. The robot is designed to be
places dumped into rivers, waterways, or reservoirs.
controlled r manually by remote control. The
The rubbish can clog the flow of water, causing
development method of this research refers to
water to become dirty and smelly so that it often
ADDIE. This method including of analysing of the
overflows and causes disasters, including flooding.
robotic cleaning system, designing the robot,
How to clean waste from water areas requires
developing the robot, implementing robot to clean
extensive resources, for example, by cleaning staff
waste in limited water areas, evaluating the
and using excavators.
effectiveness of robot in cleaning up trash for the
more extensive area. This article is focusing on the
This study aims to provide an alternative solution to
design and development of the robot.
the problem of waste in water areas by developing

© 2024 Aman George Sebastian. This is an Open Access article distributed under the terms of the Creative Commons Attribution
License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium,
provided the original work is properly credited.
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

In New Delhi’s Yamuna River, the Ro-Boat was problem. For navigation and trash cleaning on the
successfully tested as part of the Ministry of Water ground, many well-designed algorithms have been
Resources’ Ganges River Action Plan. According to developed earlier for both single robotic systems as
the engineers who created it, 50 of these systems well as for swarms . However, because of the
could clean the river in six months. difference in the dynamic environment, propulsion
system, and the difficulty to accurately determine
the current position based on relative velocity and
acceleration, these algorithms cannot be directly
used on aquatic surfaces. Also, the navigation
algorithms developed. subsection Maritime Traffic

Fig. 1: Floating wastes

Fig. 2: Maritime Traffic
The Ro-Boat technology has been recognised by
the Mas- sachusetts Institute of Technology (MIT) Water pollution is a growing problem in today’s
as one of the 20 most promising inventions to world. Our ocean is being flooded with two main
come out of India in 2013. Additionally, US-AID types of pollution: chemicals and trash.
recognised it as one of the top five water and
drainage-related innovations of 2013. Chemical contamination, or nutrient pollution, is
concerning for health, environmental, and
The Ro-Boat is still in the experimental and research economic reasons. This type of pollution occurs
stage, with a focus on boosting its carrying when human activities, notably the use of fertilizer
capacity. To expand its use, its creators are in talks on farms, lead to the runoff of chemicals into
with the Indian government. waterways that ultimately flow into the ocean. The
increased concentration of chemicals, such as
Previously some robots have been built for cleaning nitrogen and phosphorus, in the coastal ocean
water surface. A few results are available in the promotes the growth of algal blooms, which can be
open references, which have discussed the toxic to wildlife and harmful to humans. The
development of such particular purpose robot . This negative effects on health and the environment
work aims at developing a more versatile and caused by algal blooms hurt local fishing and
efficient system by the usage of the aqua robot. tourism industries.
This robot has an excellent opportunity to expand
its functions in the future. Such conducting Water trash encompasses all manufactured
activities for removing algae, leaves and twigs; products—most of them plastic—that end up in the
spraying of chemicals at the appropriate locations; ocean. Littering, storm winds, and poor waste
checking the water quality as well as deploying management all contribute to the accumulation of
payload are planned to be done autonomously. this debris, 80 percent of which comes from sources
These functions may save a lot of human effort and on land. Common types of marine debris include
provide a sustainable solution to the pervasive various plastic items like shopping bags and
beverage bottles, along with cigarette butts, bottle
2
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

caps, food wrappers, and fishing gear. Plastic waste 1. Trash in Water Bodies
is particularly problematic as a pollutant because it
is so long-lasting. Plastic items can take hundreds
of years to decompose.

This trash poses dangers to both humans and

animals. Fish become tangled and injured in the
debris, and some animals mistake items like plastic
bags for food and eat them. Small organisms feed
on tiny bits of broken-down plastic, called micro
plastic, and absorb the chemicals from the plastic
into their tissues. Micro plastics are less than five Fig. 3: Piled up Waste
millimeters (0.2 inches) in diameter and have been
detected in a range of marine species, including Mismanaged trash that escapes into waterways can
plankton and whales. When small organisms that create a wide range of problems. Aquatic trash
consume micro plastics are eaten by larger animals, affects water quality, endangers plants and animals,
the toxic chemicals then become part of their and pollutes the outdoor spaces that we depend on
tissues. for tourism and recreation. Though all types of
aquatic trash can have potentially harmful impacts,
In this way, the micro plastic pollution migrates up plastic waste is particularly concerning because of
the food chain, eventually becoming part of the its tendency to persist in the environment and its
food that humans eat. Solutions for these pollution widespread production, use, and disposal.
include prevention and cleanup. Disposable and
single-use plastic is abundantly used in today’s Plastic pollution is particularly dangerous because it
society, from shopping bags to shipping packaging does not fully biodegrade in the environment.
to plastic bottles. Changing society’s approach to Plastic pollution has been found in a wide range of
plastic use will be a long and economically organisms and habitats, including coral reefs,
challenging process. Cleanup, in contrast, may be estuaries, beaches, and the deep sea. Since plastic
impossible for some items. Many types of debris does not decompose, it continually accumulates in
(including some plastics) do not float, so they are landfills and in the environment.
lost deep in the ocean. Plastics that do float tend to
collect in large “patches” in ocean gyres. In addition to degrading the habitats and
ecosystem services that humans use, plastic aquatic
The Pacific Garbage Patch is one example of such a debris can directly interfere with navigation, impede
collection, with plastics and micro plastics floating commercial and recreational fishing, threaten health
on and below the surface of swirling ocean currents and safety, and reduce tourism. Large debris, such
between California and Hawaii in an area of about as derelict fishing nets and lines that float at or just
1.6 million square kilometers (617,763 square below the surface, pose the greatest threat to vessel
miles), although its size is not fixed. These patches navigation. Lines and nets can become wrapped
are less like islands of trash and, as the National around propellers and entrained in intakes of
Oceanic and Atmospheric Administration says, motors, and vessels may strike large items,
more like flecks of microplastic pepper swirling damaging hulls and propellers. Coast communities
around an ocean soup. Even some promising are spending more than 5 million dollars a year to
solutions are inadequate for combating marine combat litter and prevent trash from becoming
pollution. So-called “biodegradable” plastics often marine debris. This public cost burden makes yet
break down only at temperatures higher than will another reason for finding effective strategies to
ever be reached in the ocean.
3
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

reduce the trash entering our waterways and with various adjustments and practical tips to
contributing to marine debris. further enhance detection performance.

II. LITERATURE SURVEY This study will definitely help in training models
using custom training Dataset.
During the months of August, September, October
Novem- ber of 2022, a literature review for this Pond Cleaning Robot
project was completed. The literature addressing The robot in this journal removes waste materials
the advancement of automated water robots from water surfaces and disposes of them safely in
received the majority of attention. the water body. The Bluetooth-enabled pond
cleaning robot extracts trash, plastics, and waste
The literature sources that are pertinent to each from Nashik’s Godavari River. This literature journal
input are included in References. The following has pointed out the basic design and configuration
literature journals provided these source of that should be present while designing such a
information. robot.

1. Detection of Waste Materials Using Deep Arduino based Tracking System Using GPS and
Learning and Image Processing GSM
This paper is aimed at planning and developing up GPS and GSM are combined into a location tracking
a framework with a deep learning approach that system by the microcontroller. It is used to locate
can be effectively used for waste segregation. The vehicles or any other items that are connected to a
image will be recognized by utilizing the concept of tracking device using GPS. A smartphone and an
a convolutional neural network and with the help of Arduino UNO are frequently used technologies that
an image processing method that identifies wastes were effectively utilised in the suggested system.
from their shape, color, dimension, and size. This GPS is a navigation system that uses satellites to
technique automatically will help the system to deliver precise location and data. An update is sent
learn the pertinent features from the sample images and received from the object location to a database
of the trash and consequently recognize those using the GSM module.
features in new images.
SMURF
2. Automatic collaborative water surface A Fully Autonomous Water Surface Clean- ing
coverage and cleaning strategy of UAV and Robot with A Novel Coverage Path Planning
USVs Method: With cutting-edge sensors and a special
This research is based on designing an autonomous coverage path planning technique, SMURF is a
water cleaning approach and ingeniously establish completely autonomous water surface cleaning
a system architecture for the cooperation between robot. It collects and stores garbage and
UAV and USVs. From this we have to develop an contaminants and keeps track of environmental
autonomous obstacle avoidance path planning variables. SMURF is a potential method for keeping
method for our robot to enable autonomous water bodies clean and healthy because operators
navigation and cooperative cleaning. can regulate and monitor it remotely.

3. Object Detection With Deep Learning: A A Modified Yolov3 Detection Method for Vision-
Review Based Water Surface Robot
A brief overview of the development of deep This paper presents a mod- ified version of the
learning and the convolutional neural network, YOLOv3 object detection method designed for a
which serves as its paradigm. Then, we concentrate vision-based water surface garbage capture robot.
on typical generic object detection designs along It incorporates deep learning and computer vision
4
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

techniques to improve the robot’s object detection popular choice for GPS applications: compact size,
accu- racy and speed. The modified YOLOv3 low power con- sumption, serial communication,
method provides precise information aiding in the and NMEA protocol support. It is designed to
preservation of water quality and ecosystem health. operate on low power, making it suitable for
battery-powered applications where power
A Smart Autonomous Floor Cleaner with an efficiency is crucial. The NEO-6M module is a
Android- Based Controller reliable, cost-effective, and easy-to- use solution for
The development of a smart autonomous floor incorporating GPS functionality into various
cleaner with an Android-based controller aims to projects and applications.
create an intelligent cleaning robot capable of
autonomously navigating and cleaning floors while It provides accurate time information, a built-in
being controlled and monitored through an ceramic patch antenna, reliable and accurate
Android device. It utilizes a combination of positioning performance, configuration and control,
mapping, path planning, and obstacle avoidance integration and development, and multi-GNSS
algorithms to navigate efficiently and avoid support. It can achieve horizontal positioning
obstacles during cleaning. It may also include accuracy within a few meters under normal
features such as scheduling options, cleaning mode conditions and vertical accuracy within
selection, and virtual boundaries. approximately 10 meters. It can be configured and
controlled through AT commands, allowing
III. METHODOLOGY customization of various settings. Integration and
develop- ment involves connecting it to the host
1. NEO-6M Module system via serial communication and parsing the
The GPS NEO-6M module is a compact, low-power, NMEA sentences received from the module to
and cost-effective GPS (Global Positioning System) extract the desired GPS information. For
module commonly used in various applications applications requiring higher accuracy or additional
such as navigation systems, asset tracking, UAVs features, u-blox offers more advanced GPS modules
(unmanned aerial vehicles), robotics, and more. It is in their product lineup.
built on u-blox’s industry-leading
2. HC-05 Module
The HC-05 Bluetooth module is a popular and
widely used module for wireless communication
between electronic devices. It is based on the
Bluetooth 2.0+EDR specification and has a
communication range of up to 10 meters. It
communicates with the host system using a serial
interface, commonly referred to as UART, which
supports standard serial communication protocols
such as 8-N-1. The module can operate in either
master or slave mode, and can be configured and
controlled using AT commands. The range can vary
Fig. 4: NEO-6M
depending on the surrounding environment and
any potential obstructions.
GPS chipset technology, which provides high
The HC-05 Bluetooth module is a convenient and
sensitivity and excellent performance in acquiring
reliable solution for establishing wireless
and tracking GPS signals even in challenging
communication between electronic devices. Its AT
environments with weak signal strength or
commands allow users to customize settings such
interference. It offers several features that make it a
as the Bluetooth device name, pairing code,
5
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

communication baud rate, and operating mode. It data, making it compatible with various
operates on a voltage supply of 3.3V and supports microcontrollers and development boards. It offers
the Secure Simple Pairing (SSP) protocol for secure high accuracy and resolution in magnetic field
and encrypted communication. It finds applications measurements, with a typical accuracy of 2-3
in various domains, such as home automation, degrees and a resolution of up to 0.2 milligauss.
robotics, wireless sensor networks, Internet of Built- in calibration functionality compensates for
Things (IoT) projects, and more. When using the external magnetic interference or offsets, and the
HC-05 module, it is essential to ensure module provides different oper- ating modes to
compatibility and check the documentation and optimize power consumption and measurement
resources provided by the module’s manufacturer. speed.
Online tutorials and libraries are available to
simplify the integration and usage of the HC-05 Integrated features include adjustable
module in different projects. measurement range, data output rate
configuration, and automatic gain control. It is
widely supported by various libraries and code
examples avail- able for different microcontrollers
and development platforms, making it a popular
choice for projects requiring magnetic field sensing
and orientation determination.

4. A2212/6t 2200kv BLDC Motors

Fig. 5: HC-05 module

3. HMC5883L Module

Fig. 7: A2212/6t 2200kv BLDC Motors

The A2212/6T 2200KV motor is a brushless motor

used in remote-controlled (RC) aircraft and
multirotor drones. It is characterized by its specific
designations, such as ”A2212”, ”6T”, and ”2200KV”.
Fig. 6: HMC5883L It operates on brushless technology, which offers
advantages such as higher efficiency, longer
The HMC5883L module is a three-axis digital lifespan, and reduced maintenance requirements. It
compass module used for measuring magnetic is capable of delivering substantial power for its
fields. It features a three- axis magnetometer that size, and is commonly used in RC aircraft, multirotor
measures the strength and direction of magnetic drones, and other similar projects. It is compatible
fields in three dimensions (X, Y, and Z axes). Digital with a range of electronic speed controllers (ESCs).
output is provided for the measured magnetic field

6
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

The A2212/6T 2200KV motor is a popular brushless protection, and current limiting help prevent
motor known for its power and performance in RC damage to the motor and ESC. ESC 30A units often
aircraft and multirotor drone applications. It require initial configuration and calibration, and
features three wire leads, which are connected to proper installation and wiring are essential for
the ground or negative terminal, the red wire is the optimal performance and safety. Proper installation,
positive supply voltage, and the yellow or blue wire configuration, and adherence to safety guidelines
is used for signal control. Cooling and maintenance are essential for reliable and safe operation.
are important to prevent overheating during
extended operation, and regular inspection and Objective
maintenance can help ensure op- timal Location of the Robot
performance and longevity. Safety precautions GPS, localization implies more than knowing one’s
should be taken when handling the motor carefully absolute position in the Earth’s reference frame.
and following manufacturer guidelines for safe Consider a robot that is interacting with humans.
operation. This robot may need to identify its absolute
position, but its relative position with respect to
5. Electronic Speed Controllers target humans is equally important. Its localization
task can include identifying humans using its sensor
array, then computing its relative position to the
humans. Furthermore, during the Cognition step a
robot will select a strategy for achieving its goals. If
it intends to reach a particular location, then
localization may not be enough. The robot may
need to acquire or build an environmental model, a
map, that aids it in planning a path to the goal.
Once again, localization means more than simply
determining an absolute pose in space; it means
Fig. 8: Speed controller building a map, then identifying the robot’s
position relative to that map. Clearly, the robot’s
The ESC (Electronic Speed Controller) 30A is a sensors and effectors play an integral role in all the
commonly used device for controlling Brushless DC above forms of localization. It is because of the
(BLDC) motors. It serves as an intermediary inaccuracy and incompleteness of these sensors
between the power source and the BLDC motor, and effectors that localization poses difficult
receiving control signals from a compati- ble challenges. This section identifies important aspects
controller and adjusting the motor’s speed and of this sensor and effector suboptimality.
direction accordingly. It is designed to handle a
continuous current of up to 30 Amps and utilizes
Pulse Width Modulation (PWM) to control the
motor’s speed. Many ESC 30A models include a
built-in BEC, which provides a regulated output
voltage to power other components of the system.
The ESC 30A is an electronic speed controller
designed to regulate the speed and direction of
Brushless DC (BLDC) motors. It is compatible with
various types of BLDC motors, as long as the
motor’s power requirements fall within the ESC’s
current handling capabilities. Safety features such Fig. 9: Block Diagram
as low voltage protection, over-temperature

7
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

Waste Detection System goal is to imitate how the human brain functions in
order to perform some true magic.

Fig. 10: Floating Waste Detection Fig. 11: Robot Model using a Conveyor belt

Trash classification involves more various steps to Waste Collection Methods

perform. the flow steps of images are involving only There are many methods in which these floating
after image pre- processing. Image pre-processing debris can be collected and returned back to shore.
is applied in the trash dataset images to increase The first problem that is to be addressed is the
the segregation training process. the classi- fied determining the process through which the floating
features were extracted from image pre-processing waste materials will be able to enter the robot. This
using can be tackled by designing and implementing a
conveyor belt system with that can be used in our
DNN, SGD, deep learning. also, this method was environmental settings.
involved in the training process of trash
classification The type of storage that needs be created for the
robot in order to keep these waste particles is the
Deep Convolutional Neural Network next issue that arises. There are several ways to
Deep convolu- tional neural networks were the solve this issue; one is to drag the storage container
most significant technology in image processing. It or plastic mould behind the robot as it moves along
was increasing the pattern prediction intelligence of the path.
Deep learning and machine learning. DNN
algorithm has some level of complexity and usually The other solution is to mount the storage unit on
at least has two layers. this method was involved in top of the robot, which may change how much
Image Acquisition and image pre-processing. With space the robot needs
the accessibility of large amounts of data, High-
Speed GPUs, and optimized algorithms, we can IV. RESULT
now simply train computers to identify, classify and
detect the multiple categories of objects within an The Water Cleaning Robot project aimed to
image with high precision and accuracy.. It teaches develop an au- tonomous robot capable of
a computer how to identify and anticipate effectively cleaning contaminated water bodies
information by layering filters over inputs. such as lakes, rivers, and ponds. The robot was
Observations may be expressed as pictures, text, or designed to remove various types of pollutants and
audio. The way the human brain filters information debris, contributing to the preservation and
is the source of inspiration for deep learning. Its restoration of aquatic ecosystems. This project

8
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

report presents the results obtained from the performance and contribute to the continuous
development and testing of the robot. improvement of water cleaning robots.

1. Design and Construction Material Selection

 Extensive research was conducted to identify  Extensive research and consideration were
the most suitable design for an autonomous given to select suitable materials for the
water cleaning robot. construction of the robot body.
 The robot was constructed using durable and  The chosen materials were durable, lightweight,
water- proof materials to ensure its suitability and resistant to water and environmental
for aquatic environments conditions.
 It was equipped with a robust propulsion  Special attention was given to selecting
system, onboard sensors, and a collection materials that would not degrade or release
mechanism to effectively gather and remove harmful sub- stances into the water.
pollutants.

Design Process

Fig. 13: Final Framework

Fig. 12: Initial framework Conveyor Design

The water cleaning robot body developed as part of

this project demonstrated excellent performance
and functionality. The chosen materials, waterproof
construction, and ergonomic design contributed to
its durability, water resistance, maneuverability, and
ease of maintenance. The body’s compatibility with
other systems allowed for efficient integration of
pollutant detection, collection, and filtration
mechanisms. These results indicate the successful
development of a water cleaning robot body
capable of withstanding aquatic environments and Fig. 14: Conveyor design
sup- porting the robot’s mission to clean and  The design of a conveyor belt using PVC pipes
restore water bodies effectively. Further offers a cost-effective and versatile solution for
refinements and optimizations can be made based material transportation in various industries.
on these findings to enhance the body’s PVC pipes provide a durable and lightweight
framework for constructing conveyor systems.
9
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

 The next step is to select and include a belt to work together according to the instructions of
material for the conveyor such that the the user. We have been testing all these
conveyor can succesfully collect the waste components using a variable voltage supply. We
materials from the water. still need to confirm what power source to be used
in order to power the robot.
User Interface
 An app has been developed which allows the
user to control the robot via bluetooth.
 The app allows the user to create waypoints in
which the robot will be able to navigate
through, and collect the waste materials
present in that area.
 The app also allows to control the robot
manually to change direction and navigate it
via bluetooth.
 The app also allows the user get info about the
real time location of the robot.

Fig. 16: Waste Detection

We also have trained a machine learning model

using a custom dataset which can help us identify
the waste materials in the water. Development
boards that are best suited for our needs are
currently unavailabe and therefore it is not possible
to deploy our machine learning model.

V. CONCLUSION

In conclusion, autonomous river cleaning robots are

advanced robotic boats designed to autonomously
clean rivers and water bodies. They incorporate
various features and technologies such as
navigation systems, cleaning mechanisms, filtration
Fig. 15: User Application
systems, energy sources, communication systems,
and safety features. These robots operate without
We still need to fix and waterproof the motors and
human intervention, using their capabilities to
the propellers of the robot. We still need to
navigate waterways, collect debris and pollutants,
configure the code in order for all the components
filter water, and transmit data for monitoring and
to work together. We have tested each components
analysis.
individually and made sure that it works. The GPS
module helps in collecting and setting up location
The development of autonomous river cleaning
for the robot to navigate. The bluetooth module
robots has significant benefits for environmental
helps in connecting the user to the robot. We need
conservation. They con- tribute to the reduction of
to figure out the code in order for all the 4 motors
pollution and the preservation of water ecosystems
10
Aman George Sebastian. International Journal of Science, Engineering and Technology,
2024, 12:2

by efficiently removing debris and pollutants from Cleaning Robot with A Novel Coverage Path
rivers. These robots help prevent the contamination Planning Method.” Journal of Marine Science
of water sources, protect marine life, and improve and Engineering 10, no. 11 (2022): 1620.
the overall cleanliness of water bodies. 7. Li, Xiali, Manjun Tian, Shihan Kong, Licheng Wu,
and Junzhi Yu ”A modified YOLOv3 detection
By leveraging advanced technologies and method for vision-based water surface garbage
autonomous capabilities, these robots can operate capture robot.” International Journal of
continuously, covering large areas, and adapting to Advanced Robotic Systems 17, no. 3 (2020):
different river conditions. They offer a sustainable 1729881420932715.
and cost-effective solution for river cleaning, 8. Murdan, Anshu Prakash, and Pawan Kumar
reducing the reliance on manual labor and Ramkissoon. ”A smart autonomous floor
increasing the efficiency of cleaning operations. cleaner with an Android-based controller.” In
2020 3rd International Conference on Emerging
While there are still challenges to overcome, such Trends in Electrical, Electronic and
as handling complex debris or addressing specific Communications Engineering (ELECOM), pp.
local conditions, the advancement of autonomous 235-239. IEEE, 2020.
river cleaning robots holds great promise for the
future. By combining technological innovation,
environmental stewardship, and collaborative
efforts, we can work towards cleaner and healthier
rivers and contribute to a more sustainable planet.

REFERENCES

1. ”Deng T, Xu X, Ding Z, Xiao X, Zhu M, Peng K. “.

Automatic collaborative water surface coverage
and cleaning strategy of UAV and USVs. Digital
Communications and Networks. 2022 Dec 28.
2. Zhao, Zhong-Qiu, Peng Zheng, Shou-tao Xu,
and Xindong Wu. ” Object detection with deep
learning: A review.” IEEE transactions on neural
networks and learning systems 30, no. 11
(2019): 3212-3232.
3. Soumya, H. M., and Basavaraj Gadgay. ”Pond
Cleaning Robot.” International Research Journal
of Engineering and Technology (IRJET) 5, no. 10
(2018).
4. Mitra, Arghadeep. “”Detection of Waste
Materials Using Deep Learn- ing and Image
Processing.” PhD diss., California State
University San Marcos, 2020.
5. Htwe, Thin Thin, and Kyaw Kyaw Hlaing
”Arduino based tracking system using GPS and
GSM.” International Journal for Advance Re-
search and Development 4, no. 8 (2019): 11-15.
6. Zhu, Jiannan, Yixin Yang, and Yuwei Cheng
”SMURF: A Fully Autonomous Water Surface
11