VISVESHVARYA TECHNOLOGICAL UNIVERSITY

SANTHIBASTAWAD ROAD, MACHHE, BELGAUM-590014

KARNATAKA

Project Phase-I Report

on
“Robotic Fish”
Submitted in partial fulfilment of the requirements for the award of

BACHELOUR OF ENGINEERING
IN
ELECTRONICS AND COMMUNICATION ENGINEERING
Submitted by
Bharath Gowda J M 1HM18EC004
Suprith T P 1HM18EC018
Thakshak M P 1HM18EC023
Vaibhav Pol 1HM18ECO25

Under the guidance of

Dr. C P Latha M.E, Ph. D
HOD,Dept of ECE
HMSIT, Tumakuru

DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGG

H M S INSTITUTE OF TECHNOLOGY
NH-4, Kesaramadu Post, Kyathsandra, TUMAKURU-572104
2021-2022

1
 H M S INSTITUTE OF TECHNOLOGY
NH-4, Kesaramadu Post, Kyathsandra,TUMAKURU-572104
DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGG.

CERTIFICATE
Certified that the mini project report on “Robotic Fish”isabonafiedworkcarriedoutby BHARATH
GOWDA J M (1HM18EC004),SUPRITH T P (1HM18EC018),THAKSHAK M P (1HM18EC023) ,
VAIBHAV POL (1HM18EC025)in partialfulfilment for the award of degree of Bachelor of
Engineering in Electronics and Communication Engineering of the Visvesvaraya Technological
University, Belgaum during the year 2021-2022. It iscertified that all corrections / suggestions
indicated for Internal Assessment have been in thereport deposited in the departmental library.
The project phase -Ireport has been approved as it satisfiesthe academic requirements in respect
of mini project report prescribed for the degree of Bachelor of Engineering.

Dr. C P Latha
M.E, Ph. D
HOD, Dept of ECE
HMSIT, TUMAKURU

2
 ABSTRACT
Trash generated by human activities can be seen in various water environment such as lakes, ponds and
rivers. These marine trash needs to be removed as they are threat to the environment. Detecting and
collecting debris face various issues due to unique difficulties of the water environment and it can also be
used for the object detection and tracking along with collision avoidance. Autonomous vehicular
technology can be helpful to detect and collect debris in water sources. In this project, robotic fish, which is
a remotely controlled pilot less vehicle, is used to detect floating marine trash and eventually remove
them.

This intelligent robotic fish, which propels by using mechanical propellers, could help in addressing the
problem of water debris by detecting and collecting the floating trash. The system contains a garbage bin
to collect the debris. The conveyer belt, which is driven by a DC motor, is responsible for the movement of
trash foam the surface of water to the garbage bin.

Keywords:-Environment, Debris, Detection, Trash.

3
 CONTENTS

CHAPTERS PAGE NO

CHAPTER 1: Introduction 7
1.1:Motivation
1.2:Objective of the project
1.3:Organisation of the report
CHAPTER 2: Background11
CHAPTER3:Literature Survey 12
3.1: Robotic fish swimming modes
3.1.1 Propulsion using fins
3.1.2 Propulsion using mechanical propellers
3.2: Object detection using CNN
3.3: Mechanism for Debris Collection
CHAPTER 4: System Overview 17
4.1 Design Requirements
4.2 Propulsion control
4.3 Communication Module
4.4 Debris Detection
4.5 Debris Collection
CHAPTER 5: Hardware Requirements 22
5.1 ARDUINO
5.2 WI-FI MODULE (ESP32) Node MCU
5.3 LINEAR GRID RAIL MECHANISM
5.4 ESP32 CAM
5.5 12V DC motor
5.6 L293D motor driver
5.7 12v Battery
CHAPTER 6: Software Requirements 29
6.1Arduino IDE software
6.2 Blynk Application Software

4
CHAPTER 7:Proposed Advantages33

CHAPTER 8: Proposed Applications 34

CHAPTER 9: Conclusion 35
9.1 Future Enhancement

CHAPTER 10: Refrences 37

5
List of Figures

3.1 Parts of a 3 blade propeller. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

3.2 CNN Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3.3 Conveyor System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.1 Block Diagram. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.2 Forces and moments for marine propeller. . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.3 Remote Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.4 Esp32 cam directly provides the live video streaming from the Robot fish to designed
app. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.5 Conveyor belt system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
5.1 Arduino Board. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
5.2 Wi-Fi Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5.3 Grid Mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5.4 ESP32 Camera Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
5.5 L293D Driver module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
5.6 Battery. . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6.1 Arduino IDE software icon. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.2 Blynk communication network. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

6
CHAPTER 1

Introduction
In recent years, bioinspired underwater vehicles have become a significantly hot research
topic in the field of ocean engineering. Especially, fish-like robots have attracted the
attention of the research community because of its great advantages over conventional
propeller-driven underwater robots, such as high efficiency, extreme swiftness, and station-
holding ability.1 Furthermore, fish robots have shown better performance in terms of
swimming mechanism. Two Such kinds of robots have a wide range of underwater
applications and have various other applications rather than deep-sea scientific exploration,
military, advertisement, and entertainment. The fish robot has a more advanced feature in
terms of speed, sensing, and intelligence as compared to real sea creatures. For instance, due
to advanced vision technology, now underwater robots have a better ability to identify the
targets and obstacles in the aquatic environment.

Accumulation of waste in water bodies adversely effect the marine eco-systems and pose a
longterm threat to aquatic species. This vast amount of trash in the water source that injures
aquatic life and contaminates the water must be removed periodically. An effective strategy
for removing debris from marine environment is the use of autonomous under water vehicles
for trash detection and removal.

An autonomous under water vehicle is a robot that travels on the surface and under water
which can be used to conduct under water researches. Remotely operated under water
vehicle (ROV) is a unoccupied vehicle, which is fitted outwith sensors and others tools to
collect various types of data. These vehicles provide a safe alternative to explore dangerous
areas. RIOV’S equipped with many tools,have applications in many fields including military
and science.

To attain better and efficient manoeuvrability of the fish robot, it is necessary to have
information about various parameters of the fluid in which the fish robot is swimming. These
parameters include fluid flow and pressure that the fluid exerts on the fish robot.
Computational fluid dynamics (CFD) designed tool is widely used to predict fluid flow in
7
different domains. It also measures the drag force faced by the fish robot when moving in the
fluid. Besides, it also plays a vital role to test the prototype model prior to the actual design
created.

The water pollution is very important problem in rivers, ponds and water bodies near River
. Due to increase in water pollution in the form to waste debris; it is hampering the life of
aquatic animal and make their life in danger. Similarly sometimes the aquatic animal tends
to eats surface waste debris considering it as a food; which ultimately cause the death of
animals. Due to polluted water is are many skin diseases to human kind are observed. So
that to reduce the water pollution we are trying to make pond cleaning robot.“Robot Fish'' a
device which involves the removing the waste debris from water surface and safely

dispose from the water body. The Robot Fish works on Bluetooth to extract waste water
debris, plastics & garbage from River.

The biggest impact of cleaning the chemical wastes can cause respiratory diseases and it
plays a challenging issue for the municipality officers Water damage is classified as three
types of contaminated water. They are clean water, gray water and black water. Clean water
is from a broken water supply line or leaking faucet. If not treated quickly, this water can
turn into black water or gray water, depending on length of time, temperature, and contact
with surrounding contaminants.

Drainage pipes are used for the disposal of sewage and unfortunately sometimes there may
be loss of human life while cleaning the blockages in the drainage pipes. The municipality
workers are only responsible to ensure that the sewage is clean or not. Though they clean the
ditches at the side of buildings, they can’t clean in very wide sewages. The municipality
workers need to get down into the sewage sludge to clean the wide sewage. It affects their
health badly and also causes skin allergies.

8
1.1 Motivation
Trash in various types of water bodies is common. Removal of these pollutants in small
water reservoirs like water tanks and swimming pools has to be done frequently. Plastic
waste are the most common type of debris found on the surface of water.Robotic
vehicles make this task much easier.
Trash enters water bodies in various ways and contaminates the water. Maintaining
clean water is essential and involves the removal of debris more frequently. Out of
various types of marine trash, plastic wastes are the most dominant waste material
found on the water surface. Recycling and other activities to keep debris out of the
water reservoirs are also very important but there is a need for an efficient strategy to
remove the debris which are already in the water resource. One of the widely used
techniques for collecting plastic from water source is refuse-removal vessel method.
However, this method is appropriate only to large rivers with a significant accumulation
of plastic and cannot be implemented to clean small water sources with low-density
waste. Hence, a robotic system may be helpful to clean small water bodies like water
tanks, swimming pools, small ponds, etc

1.2 Objective of the project

There are mainly three objectives in the project. These objectives are listed below.
1. To build a robot that moves smoothly on the surface of the water.
2. To remotely control the robotic fish using Radio Frequency (RF) technology.
3. To make the robot capable of finding and collecting certain kinds of trash like small
plastic covers and bottle caps floating on the surface of water.

The process of detecting the floating trash visually in natural water environments is
achieved using deep learning algorithms. The convolutional neural network is one of
the widely used deep learning networks for analyzing, processing and classifying
visual data. CNN algorithm is used to achieve the trash detection task. The system

9
 consists of a garbage bin to collect the trash. The conveyor system efficiently
transfers the floating trash into the garbage bin.

1.3 Organisation of the report

The project report is presented in a number of chapters, Chapter 1 comprises of

introduction, motivation and objective of the project. Chapter 2 consists of literature
review. Chapter 3 gives the information about the system overview. System hardware and
System software are presented in Chapter 4 and chapter 5 respectively. Chapter 6 consists
of details of experiments and results. Finally, the conclusion and scope for future work are
presented in Chapter 7.

10
CHAPTER 2

Background

Bionic or biologically inspired engineering is covering all the fields and there has been a
rapid increase in research and development of the bio-inspired mechatronic system.
Robotic fish is an example of bio inspired AUV’s and it is found that robotic fish performs
better and are more competent than current AUV’s in the field of underwater explorations
[1]. Since 1993, the unmanned surface vehicles has been evolving for various missions such
as navigation system, water environment exploration, scientific researches on aquatic
species and various military applications. Robotic fishes have better propulsion capabilities,
stability, efficiency and exhibits their vast potentials for researches. Robotic technology is
making exploration cheaper and more accessible with its ability to withstand harsh
environments, tides and ocean currents [2]. Safety is the main advantage of utilizing
robotics so that there will be no risk to human life. Robotic fish can record videos and take
pictures. The robot can also gather information about artifacts and biological environment
of the water body.

11
CHAPTER 3

Literature Survey

Mohammed Idris [1] proposed system, the machine is operated with remote control to clean
the sewage. The system has a wiper motor that starts running as soon as the set-up is
switched on. Two power window motors are connected to the wheel and it is driven with
the help of the remote control set-up. The process starts collecting the sewage wastes by
using the arm and it throws back the waste into the bin fixed in the machine at the bottom.
An arm is used to lift the sewage and in turn a bucket is used to collect them. The set up
runs even in sewage area with water (limited to a particular amount) so that the wastages
which floats on the water surface also gets collected.

Mr. Abhijeet.M. Ballade proposed system explains that, Due to increase in water pollution
in the form to waste debris; it is hampering the life of aquatic animal and make their life in
danger. So that to reduce the water pollution we are trying to make river cleanup
machine.“River cleanup machine'' a machine which involves the removing the waste debris
from water surface and safely dispose from the water body. The river cleanup machine
works on hydropower to extract waste water debris, plastics & garbage from water.

Mr. P. M. Sirsat proposed a paper emphasis on design and fabrication details of the river
waste cleaning machine. This machine has designed to clean river water surface. The remote
operated river cleaning machine has designed which helps in river surface cleaning
effectively, efficiently and eco-friendly. The “River waste cleaning machine” is used where
there is wastedebris in the water body which are to be removed. This achine consists of DC
motors, RF transmitter and receiver, propeller, PVC pipes and chain drive with the conveyor
attached to it for collecting wastage, garbage & plastic wastages from water bodies.

12
3.1 Robotic fish swimming modes
Generally, there are two ways of designing robots swimming on the water surface.
The first technique employs fish like swimming modes which propels using fins attached to
it. The second method is more like a robotic boat equipped with mechanical propellers for
smooth motion.

3.1.1 Propulsion using fins

Generally, a robotic fish exhibits undulatory motion in either of the two ways:
1. Using the body and/or caudal fin (BCF) movements.
2. Using median and/or paired fin (MPF) movements.

BCF provides greater thrust and accelerations. Robotic fish depends on undulation
movements to produce the main propel energy. This type of propulsion is less noisy,
more effective and maneuverable.
The swim patterns of fish can be periodic or steady or unsteady (includes, fast start,
sharp turns and brakes). Basic swim patterns can be divided as follows [3]:

1. Straight motion: The fish moves straight at a constant speed.

2. Turning: Fish turns at a small angular speed with constant linear speed.
3. Coast: Fish body is kept stationary.
4. Brake: The fish imparts a sudden de-acceleration.

All the swim patterns are generated by special tail motion and pectoral fin motion.
Caudal/tail fin is used to produce propulsion. Pectoral fins are found in pairs which
are used to maintain depth in water, turn and stop quickly [4].

13
3.1.2 Propulsion using mechanical propellers

Underwater vehicles achieve transmission with thrusters and propellers. These

mechanical devices drive the underwater robot forward or backward which depends
on the direction of rotation or pitch of the propeller [5].
Propellers are attached to the robot body via shaft arrangement. As shown in Figure
2.1, propellers consist of radiating blades which when rotated exerts linear thrust on
the water. The blades are shaped in such a way that their rotational motion in the
water creates a pressure difference between the two surfaces of the blade.

Figure 3.1: Parts of a 3 blade propeller.

As motor rotates the propeller, rotational motion is converted into thrust (linear in
nature). The resulting force imparts momentum to the water and pushes the robot
forward. In order to have backward motion, the motor and hence the propeller is
rotated in opposite direction so that the thrust is reversed.

3.2 Object detection using CNN

The field of computer vision has been going through tremendous developments as
the technology has enormous applications such as scene understanding, robotics,
video surveillance, and self-driving systems [6]. Visual recognition systems, which is
the core for all these applications, includes image classification, localization and
14
detection. As significant developments in neural networks has been achieved, the
visual recognition systems have reached exceptional performance. Object detection
is one among these domains witnessing great success in computer vision.
Convolutional neural network can be considered as a class of deep learning network.
It employs feed-forward approach. This algorithm has been efficiently used to
produce accurate results in computer vision tasks, such as image classification and
detection [7]. CNNs are like traditional neural network, but they are composed of
deeper layers. Figure 2.2 shows the CNN architecture. It is made up of three
important layers namely, convolutional layer, pooling layer and fully connected
layer. Convolutional layer and pooling layer are alternated and hence the depth of
each filter keeps on increasing from left to right and the output size (height and
width) decreases eventually. The fully connected layer is the final stage in the
network which is similar to the last layer of the conventional neural networks. The
input to the model is an image that consists matrix of pixel

Figure3.2:CNNArchitecture
values. A neuron in CNN is similar to biological neurons and produces the sum of its
input values. In convolutional layer, neurons are connected to the local regions in
the image. The convolutional layer’s parameters are made up of a set of learnable
filters, also called as kernels. These filters are convolved across the width and height
of the input volume. Scalar product of the input values and the filter values is
computed. This process Dept.of E&CE, S.I.T.,Tumakuru-03 5 Robotic Fish 2020-21
results in a 2-dimensional activation map of that kernel. In this way the network
learns the filters. After the model learns about the features, if it finds some particular
type of feature at some spatial position in the input frame, a trigger is initiated. This
is how the convolutional neural network learns and detects objects in the input
15
images.

3.3 Mechanism for Debris Collection

Robotics has seen various applications in industrial automation as robotic

manipulators and automatic guided vehicles have proved to improve efficiency in
industrial automation applications. A robotic arm is one of the widely used efficient
and preferred electromechanical machines for many applications in most of the
industrial and dangerous areas [8]. Most of the floating debris are small and
deformable. Therefore, it is ineffective to use robotic arms or manipulators to pick
the trash one by one. Conveyor systems are reliable, durable and less expensive
component to move or carry various things. This system is labor saving and used in
automated distribution and warehousing. The system consists of two or more
pulleys. The belt rotates about them and forms a closed loop [10]. Figure 2.3 shows
the conveyor system in which pair of tongs helps to move the trash to the front of the
conveyor belt. The belt lifts the trash and dumps it in the garbage bin. The conveyor
belt mechanism has the advantages of being simple, easy to control, suitable for
collecting different types of waste and cost effective.

Figure 3.3: Conveyor System.

16
CHAPTER 4

System Overview

The project can be divided into mainly three parts. First part is the robot design
which includes motion control of the robot. Second part is to detect the debris and
the third part is to collect the debris based on the detection results. The overall block
diagram is show in Figure 3.1.

BATTERY

MOTOR DRIVERS FOR

ARDUINO UNO ROBOT FISH
MOTOR DRIVERS FOR MOVEMENT
ROBOT FISH
MOVEMENT M1 M2

M3 M4
MOTOR MOTOR
DRIVER 1 DRIVER 2

LINEAR LINEAR
GRID RAIL GRID RAIL
MECHANIS MECHANIS
M FOR FOR
CONVERY CONVERY

Android phone
WIFI MODULE ESP32 CAM Wi-Fi
module

Figure 4.1: Block Diagram.

17
The block diagram of Robotic is shown in the abovefigure. The robot is controlled through
IoT. Android App is designed whichcommunicates with the Arduino microcontroller board
based on the
ATmega via ESP32 wi-fi module through wi-fi. The Robot has DC geared
motors along with L298N motor driver where the propellers are attached to shaft of the
DC geared motors. The motor driver (L298N) is used to activate the moving of the
gear motor, where the movement of the robot is controlled by an Android app.

The ESP32- CAM with OV2640 Camera module is used for live video streaming of the
robot. The ESP32-wifi module acts as an interface between Android app and the
Microcontroller. The Microcontroller will be programmed with the help of the
Embedded C programming. The ESP32 Wi-Fi module used to control the movement
of the robot and control the direction of the camera rotation. In this proposed system, a
rechargeable battery is used to supply power to the electronic components of the
system.

Along with that another two DC motors are used as a converyer belt purpose to take the
waste from the pond/ocean/sea

4.1 Design Requirements

Several factors and requirements are to be considered to design the robot and achieve
trash cleaning in small tanks and swimming pools. Major design requirements are listed
below:

• Should be capable of exhibiting undulatory motion in the small water reservoirs.

• Should light in weight be and compact in size .

• System should be able to collect various kinds of small sized trash.

18
• Should be capable of working continuously for 3-4 hours after fully charged.

4.2Propulsion control

A pair of mechanical propellers connected to either side of the robot are responsible for the
movement of the robot on the surface of water. Propellers are driven by 60 rpm DC motors.
As the motors rotate, propellers also rotate and the blades exerts linear thrust on the
water. This leads to pressure difference between the surface in front and back of the
propeller. As a result, water is accelerated in one direction creating a reactive force which
makes the body attached to the propellers to move. Similarly, for backward movement, the
propellers are made to rotate in opposite direction and hence, reverse thrust is applied
which leads to backward movement. In case of left and right turn, one of the two propellers
is made stationary and other propeller is rotated. Hence, displacement of water occurs only
on one side which results in the turning of the body. Figure 3.2 depicts the forces acting on
the propeller blades in a 3 blade propeller. DC motors are the electrical devices that
transforms electrical energy into mechanical energy. The DC motor driver acts as an
interface between the motors and the control circuits.

4.3 Communication Module

Robotic fish is remotely controlled using radio technology. The block diagram of the emote
module is shown in Figure 3.3. There are four switches on the remote module to control
the movement of the fish. Whenever one of these switch is pressed by the user, the
corresponding information is transmitted at a frequency of 2.4GHZ through antenna. The
radio receiver receives the transmitted message through its antenna. Based on the

19
Figure 4.2: Forces and moments for marine propeller.

control information required action is taken to change the direction of the robot. Four
control switches to change the direction of the robot are listed below.

• Forward

• Backward

• Left

• Right

Figure 4.3:RemoteModule.

4.4 Debris Detection

The input video from robotic fish is converted into frames. Frame is given as input to
convolution neural network classifier to detect the debris in the image. A CNN works by
extracting features from images. This eliminates the requirement for manual feature

extraction. As the network trains on image dataset, the features are learnt by the model.
This makes deep learning models highly accurate and suitable for computer vision tasks.
CNN learns feature detection through several hidden layers. Each layer increases the
complexity of the learned features. Dataset of actual debris in open-water locations is used

20
for training CNN based detector for object detection in real time. The trained network is
then evaluated on test dataset that are not used for training the network.

4.4Esp32 cam directly provides the live video streaming from the Robot fish to designed app.

4.5 Debris Collection

Conveyor belt mechanism is used to collect the floating debris. Conveyor belt brings small
sized trash which is present on the surface of water into the garbage bin. Conveyor system
consists of a head pulley, a tail pulley and mesh-like belt as shown in Figure 3.4. A DC motor
drives the head pulley. Mesh-like belt is used in order to reduce weight. During the process
of collecting trash, mesh-like structure prevents water entering the bin.

21
Figure4.5:Conveyorbeltsystem.

CHAPTER 5
HARDWARE REQUIREMENTS

5.1 ARDUINO
5.2 WI-FI MODULE (ESP32) Node MCU
5.3 LINEAR GRID RAIL MECHANISM
5.6 ESP32 CAM
5.7 12V DC motor
5.6L293D motor driver
5.7 12v Battery

Description of hardware components used in the project is given in detail.

5.1 Arduino Uno

Arduino UNO R3 SMD is the open source Embedded Development board launched by
Arduino based on Atmega328 SMD Package Microcontroller. Because Atmel is moving
more and more of their production capacity to surface mount ICs, the DIP packaged ATmega
is becoming more and more difficult to get. To keep up with demand, we now offer the
Arduino Uno R3 with an SMD ATmega. The board is identical to the PTH version of the
Uno, but you won't be able to remove the ATmega without some hot-air. This change
shouldn't affect most users.

22
5.1 Arduino Board

Specifications:-

 Microcontroller : ATmega328 SMD.

 Operating Voltage : 5V.
 Supply Voltage recommended : 7-12V DC.
 Digital I/O Pins : 14 (of which 6 provide PWM output).
 Analog Input Pins : 6.
 DC Current per I/O Pin : 40 mA.
 DC Current for 3.3V Pin : 50 mA.
 Flash Memory : 32 KB (ATmega328) of which 0.5 KB used by bootloader.
 SRAM : 2 KB (ATmega328).
 EEPROM : 1 KB (ATmega328).
 Clock Speed : 16 MHz.

5.2 Wi-Fi module: WEMOS D1 MINI:-

Wemos D1 mini is like a "little Arduino with wifi" for a great price. It's based around the
ESP32, has one analogue port and 11 digital ports. It's programmed via micro-USB (or
remote flash via wifi). You can use it with the Arduino IDE, micropython or NodeMCU. It
runs from 5V or 3.3V

23
 5.2 Wi-Fi Module

Specification and Features :-

 11 digital input/output pins, all pins have interrupt/pwm/I2C/one-wire supported(except D0)

 1 analog input(3.2V max input)
 Micro USB connection
 Use BRT (Bias Resistor Transistor), easier into flash mode.
 30% increase in radiating area, more stable.
 Dimensions : 35 x 26 x 8 (LxWxH) mm.
 Weight: 10 gm.
 Board Color : Blue
 Compatible with Arduino

5.3 LINEAR GRID RAIL MECHANISM

5.3 Grid mechanism

24
Linear blocks connect to the moving components by mounting to the block to
support the motion for the specified component. The motion is created along the rail
through recirculating ball bearings within the block that is contacted on the rail. Two
basic types of linear guides are predominantly used in industry today.

5.4 ESP32 CAM

5.4 ESP32 cam module

The ESP32 CAM WiFi Module Bluetooth with OV2640 Camera Module 2MP For Face
Recognization has a very competitive small-size camera module that can operate
independently as a minimum system with a footprint of only 40 x 27 mm; a deep sleep
current of up to 6mA and is widely used in various IoT applications.

It is suitable for home smart devices, industrial wireless control, wireless monitoring, and
other IoT applications.

This module adopts a DIP package and can be directly inserted into the backplane to realize
rapid production of products, providing customers with high-reliability connection mode,
which is convenient for application in various IoT hardware terminals.

ESP integrates WiFi, traditional Bluetooth, and BLE Beacon, with 2 high-performance 32-
bit LX6 CPUs, 7-stage pipeline architecture. It has the main frequency adjustment range of
80MHz to 240MHz, on-chip sensor, Hall sensor, temperature sensor, etc.

Features :
ESP32-CAM :

1. The smallest 802.11b/g/n Wi-Fi BT SoC module.

2. Low power 32-bit CPU, can also serve the application processor.
3. Up to 160MHz clock speed, summary computing power up to 600 DMIPS.
25
 4. Built-in 520 KB SRAM, external 4MPSRAM.
5. Supports UART/SPI/I2C/PWM/ADC/DAC.
6. Support OV2640 and OV7670 cameras, built-in flash lamp.
7. Support image WiFI upload.
8. Supports TF card.
9. Supports multiple sleep modes.
10. Embedded Lwip and FreeRTOS.
11. Supports STA/AP/STA+AP operation mode.
12. Support Smart Config/AirKiss technology.
13. Support for serial port local and remote firmware upgrades (FOTA).

Specifications :

1. Wireless Module: ESP32-S WiFi 802.11 b/g/n + Bluetooth 4.2 LE module with PCB
antenna, u.FL connector, 32Mbit SPI flash, 4MBit PSRAM.
2. External Storage: micro SD card slotup to 4GB.
3. Camera

a) FPC connector.
b) Support for OV2640 (sold with a board) or OV7670 cameras.
c) Image Format: JPEG(OV2640 support only ), BMP, grayscale.
d) LED flashlight.

4. Expansion: 16x through-holes with UART, SPI, I2C, PWM.

5. Misc: Reset button.
6. Power Supply: 5V via pin header.
7. Power Consumption.
a) Flash LED off: 180mA @ 5V.
b) Flash LED on to maximum brightness: 310mA @ 5V.
c) Deep-sleep: 6mA @ 5V min.
d) Modem-sleep: 20mA @ 5V min.
e) Light-sleep: 6.7mA @ 5V min.
8. Dimensions (ESP32): 40 x 27 x 12 (LxWxH) mm.
9. Temperature Range: Operating: -20 ~ 85 ; storage: -40 ~ 90 @ <90%RH.

26
5.5 12V DC motor
DC Motor – 10RPM – 12Volts geared motors are generally a simple DC motor with a
gearbox attached to it. This can be used in all-terrain robots and variety of robotic
applications. These motors have a 3 mm threaded drill hole in the middle of the shaft thus
making it simple to connect it to the wheels or any other mechanical assembly.

10 RPM 12V DC geared motors widely use for robotics applications. Very easy to use and
available in standard size. Also, you don’t have to spend a lot of money to control motors
with an Arduino or compatible board.

Specifications and Features:-

 RPM: 10.
 Operating Voltage: 12V DC
 Gearbox: Attached Plastic (spur)Gearbox
 Shaft diameter: 6mm with internal hole
 Torque: 7 kg-cm
 No-load current = 60 mA(Max)
 Load current = 300 mA(Max).

5.6 L293D MOTOR DRIVER

This L293D driver module is a medium power motor driver perfect for driving DC Motors
and Stepper Motors. It uses the popular L293D motor driver IC. It can drive 4 DC motors in
one direction, or drive 2 DC motors in both the directions.

5.5 L293D Driver module

27
Features of L293D Motor Driver Board:-

 L293D motor driver IC.

 Male burg stick connectors for supply, ground and input connection
 Screw terminal connectors for easy motor connection
 On Board LM7805 Voltage Regulator

5.7 12v battery

DC batteries use direct current, which flows in a single direction and is
generally used to power small appliances, radios, laptops, mobile phones and other
electronic gadgets.

5.6 12v Battery

28
CHAPTER 6
Software Requirements

6.1ARDUINO IDE SOFTWARE

6.2 Blynk Application software

6.1 ARDUINO IDE SOFTWARE:

The program code written in Arduino IDE is known as a sketch.The Arduino IDE software
used for developing sketches for WemosD1mini.

This IDE contains the following parts in it.

 Text editor: This is where the interpreted code can be written using a
simplified version of C++ programminglanguage.
 Message section: It shows error and also gives a feedback on saving
andexporting thecode.
 Text: The soothe displays text output by the Arduino environment along
with complete error messages and otherinformation

29
 Console Toolbar: This toolbar encompass various buttons like Verify,
Upload, New, Open, Save and Serial Monitor. On the bottom right hand corner
of the window there displays the Development Board and the Serial Port inuse.

Fig.6.1: Arduino IDE software icon

30
Fig.6.2: Arduino IDE Sketch book

31
6.2 BLYNK APPLICATION

Blynk is a Platform with IOS and Android apps to control Arduino, Raspberry Pi and the
likes over the Internet. It's a digital dashboard where you can build a graphic interface for
your project by simply dragging and dropping widgets.

Blynk was designed for the Internet of Things. It can control hardware remotely, it can
display sensor data, it can store data, vizualize it and do many other cool things.
There are three major components in the platform:
 Blynk App - allows to you create amazing interfaces for your projects using various
widgets we provide.
 Blynk Server - responsible for all the communications between the smartphone and
hardware. You can use our Blynk Cloud or run your private Blynk server locally. It’s
open-source, could easily handle thousands of devices and can even be launched on a
Raspberry Pi.
 Blynk Libraries - for all the popular hardware platforms - enable communication with the
server and process all the incoming and outcoming commands.

Fig.6.3 Blynk communication network

Features

32
 Similar API & UI for all supported hardware & devices

 Connection to the cloud using:

o WiFi
o Bluetooth and BLE
o Ethernet
o USB (Serial)
o GSM
o LCD
o LED
 Set of easy-to-use Widgets
 Direct pin manipulation with no code writing
 Easy to integrate and add new functionality using virtual pins
 History data monitoring via SuperChart widget
 Device-to-Device communication using Bridge Widget
Sending emails, tweets, push notifications, etc.

33
CHAPTER 7
PROPOSED ADVANTAGES:

 It benefits with flexible manoeuvrability(quality of moving freely) and which can

remove contaminants from water.

 Underwater Exploratory Robots can go underwater longer and deeper than humans,
they can take samples and test water, they can travel in water that is not suitable for
humans, they are used for research about animals and underwater wildlife.

 Safety is the most obvious advantage of utilizing robotics. Heavy machinerythat runs
at hot temperatureand sharp objects can easily injure a human being. By delegating
dangerous tasks to a robot, you’re more likely to look at a repair bill than a serious
medical bill or a lawsuit. Employees who work dangerous jobs will be thankful that
robots can remove some of the risks.

 Speed: Robots don’t get distracted or need to take breaks. They don’t request vacation
time or ask to leave an hour early. A robot will never feel stressed out and start
running slower. They also don’t need to be invited to employee meetings or training
session. Robots can work all the time, and these speeds up production. They keep your
employees from having to overwork themselves to meet high pressure deadlines or
seemingly impossible standards.

 Consistency:Robots never need to divide their attention between a multitude of

things. Their work is never contingent on the work of other people. They won’t have
unexpected emergencies, and they won’t need to be relocated to complete a different
time sensitive task. They’re always there, and they’re doing what they’re supposed to
do. Automation is typically far more reliable than human labour.

 Perfection: Robots will always deliver quality. Since they’re programmed for precise,
repetitive motion, they’re less likely to make mistakes. In some ways, robots are
simultaneously an employee and a quality control system. A lack of quirks and
preferences, combined with the eliminated possibility of human error, will create a
predictably perfect product every time.

34
CHAPTER 8
PROPOSED APPLICATIONS:

 Fish robots are widely used in,

1. ocean investigations
2. military operations
3. marine environment protection.

 It can also be used to locate undiscovered under water bodies.

 To identify Underwater objects.

 And also, in Sewage cleaning process.

 It can be used to detect the unidentified human bodies in aquatic areas.

35
CHAPTER 9

Conclusion
• Water surface cleaning robot is developed and tested in an experimental tank setup.
• The robot has undulating motion on calm water surface.
• Robot is capable of detecting like water bottles, plastic caps and plastic bags floating
on the water.
• The model collects small sized trash like small plastic covers and bottle caps
• A robotic fish for water surface cleaning task is developed and tested in an
experimental water tank setup.
• The robot, with the use of mechanical propellers, has an undulating motion on calm
water surface.
• A remote module, which uses the radio frequency technology, is used to control the
movement of the robot system.
• Four control switches namely, forward, backward, right turn and left turn, are used to
control the movement of the system.
• The robot is capable of detecting trash like water bottles, plastic caps and plastic bags
floating on the water.
• The model collects small sized trash like small plastic covers and bottle caps and is
dumped into the garbage bin.
• The detection of trash is achieved through YOLO model which is based on deep
neural network. The conveyor system aids in collecting the trash.
• The developed model can be useful in cleaning water bodies like water tanks and
swimming pools.

36
9.1 Future Enhancement

1. The motion of the robot is manually controlled. Obstacle avoidance mechanism can be
implemented to make the robot autonomous.

2. The detection and collection mechanism can be extended to more variety of trash.

3. The project is confined to detection and collection of floating water debris. The problem
of submerged marine debris can be addressed by making the robot capable of diving inside
the water.

37
CHAPTER 10

References
[1]. M. Mohamed Idhris, M.Elamparthi,C. Manoj Kumar, Dr.N.Nithyavathy, Mr. K.
Suganeswaran, Mr. S. Arunkumar,“Design and fabrication of remote controlled sewage
cleaning Machine”, IJETT – Volume-45 Number2 -March 2017

[2]. Mr.Abhijeet.M. Ballade, Mr. Vishal.S. Garde, Mr.Akash.S. Lahane and

Mr.Pranav.V.Boob, “Design & fabrication of river cleaning system”, IJMTER Volume
04, Issue 2, [February– 2017] ISSN (Online):2349–9745.

[3]. Mr. P. M. Sirsat, Dr. I. A. Khan, Mr. P. V. Jadhav, Mr. P. T. Date, “Design and
fabrication of River Waste Cleaning Machine”, IJCMES 2017 Special Issue-1 ISSN:
2455-5304

[4]. Information on earths-water. Accessed on Dec 4 2019 [online]. Available:

https://www.ngwa.org/what-is-groundwater/Aboutgroundwater/information-on-earths-
water

[5]. River pollution and ganga cleaning. Accessed on Dec 4 2019 [Online].Available:
https://www.businessinsider.in/science/see-photos-of-thedevastating-pollution-in-indias-
holy-ganges river/article show/62684561.cms

[6]. Su, C., Dongxing, W., Tiansong, L., Weichong, R., & Yachao, Z. (2009). An
Autonomous Ship for Cleaning the Garbage Floating on a Lake. 2009 Second
International Conference on Intelligent Computation Technology and Automation.
doi:10.1109/icicta.2009.579

38
[7]. H. Albitar, A. Ananiev and I. Kalaykov, "New concept of inwater surface cleaning
robot,"2013 IEEE International Conference on Mechatronics and Automation,
Takamatsu, 2013, pp. 1582-1587, doi: 10.1109/ICMA.2013.6618150.

[8]. Aakash Sinha, Prashant Bhardwaj, Bipul Vaibhav, and Noor Mohommad "Research and
development of Ro-boat: an autonomous river cleaning robot", Proc. SPIE 9025,
Intelligent Robots and Computer Vision XXXI: Algorithms and Techniques, 90250Q

[9]. Hu Y, Wang L, Liang J, et al. Cooperative box-pushing withmultiple autonomous robotic

fish in underwater environment.IET Control Theory Appl 2011; 5: 2015–2022.
[10]. Na K-I, Jeong I-B, Han S, et al. Target following with a visionsway compensation for
robotic fish Fibo. In: 2011 IEEEinternational conference on robotics and biomimetics,
KaronBeach, 7–11 December 2011, pp. 2114–2119. IEEE.
[11]. Hu YH, Zhao W, Xie GM, et al. Development and targetfollowing of vision-based
autonomous robotic fish. Robotica2009; 27: 1075–1089.
[12]. Wang W and Xie GM. Online High-precision probabilisticlocalization of robotic fish
using visual and inertial cues.IEEE Trans Ind Electron 2015; 62: 1113–1124.
[13]. Zhou C, Tan M, Gu N, et al. The design and implementationof a biomimetic robot
fish. Int J Adv Robot Syst 2008; 5:185–192.
[14]. Niu XL, Xu JX, Ren QY, et al. Locomotion learning for ananguilliform robotic fish using
central pattern generatorapproach. IEEE Trans Ind Electron 2014; 61: 4780–4787.
[15]. Kodati P, Hinkle J, Winn A, et al. Micro autonomous roboticostraciiform (MARCO):
hydrodynamics, design, and fabrication. IEEE Trans Robot 2008; 24: 105–117.
[16]. Yu JZ, Wang C, and Xie GM. Coordination of multiplerobotic fish with applications to
underwater robot competition. IEEE Trans Ind Electron 2016; 63: 1280–1288.
[17]. Park Y-J, Jeong U, Lee J, et al. Kinematic condition formaximizing the thrust of a
robotic fish using a compliantcaudal fin. IEEE Trans Robot 2012; 28: 1216–1227.
[18]. Chai X, Gao F, Qi C, et al. Obstacle avoidance for a hexapodrobot in unknown
environment. Sci Chi Technol Sci 2017; 60:818–831.

39
[19]. Al-Khatib H, Antonelli G, Caffaz A, et al. Navigation, guidance and control of
underwater vehicles within the widelyscalable mobile underwater sonar technology
project: anoverview. IFAC-PapersOnLine 2015; 48: 189–193.
[20]. Foresti GL. Visual inspection of sea bottom structures by anautonomous underwater
vehicle. IEEE Trans Syst ManCybern B 2001; 31: 691–705.

40