Heaven’s light is our guide

Rajshahi University of Engineering and Technology, Bangladesh

Department of Mechanical Engineering

A thesis report on
Design, Development and Wireless Control of a Flexible Robotic
Arm.

Supervised by Submitted by

Dr. Md. Rokunuzzaman Md. Asifur Rahman Sakib

Professor Roll No.: 1902140
Department of Mechanical Engineering Samit Sharma
Rajshahi University of Engineering and Technology Roll No.: 1902174

June 2025

i|Page
 Heaven’s light is our guide

Rajshahi University of Engineering and Technology, Bangladesh

Department of Mechanical Engineering

A thesis report on
Design, Development and Wireless Control of a Flexible Robotic Arm.

This report is submitted in partial fulfilment of the requirements for the degree of Bachelor of
Science in Mechanical Engineering at Rajshahi University of Engineering and Technology,
Rajshahi, Bangladesh.

Supervised by Submitted by

Dr. Md. Rokunuzzaman Md. Asifur Rahman Sakib

Professor Roll No.: 1902140
Department of Mechanical Engineering Samit Sharma
Rajshahi University of Engineering and Technology Roll No.: 1902174

ii | P a g e
 Heaven’s light is our guide

Rajshahi University of Engineering and Technology, Bangladesh

Department of Mechanical Engineering

Certificate

This is to certify that the thesis entitled “Design, Development and Wireless Control of a
Flexible Robotic Arm” has been carried out by Md. Asifur Rahman Sakib, Roll: 1902140 and
Samit Sharma, Roll: 1902174 under my supervision in the Department of Mechanical Engineering
at Rajshahi University of Engineering and Technology, Bangladesh.

Supervised by Signature of the students

……………………….. …………………………
Md. Asifur Rahman Sakib
Dr. Md. Rokunuzzaman Roll No.: 1902140
Professor
Department of Mechanical Engineering ………………………....
RUET, Rajshahi-6204 Samit Sharma
Roll No.: 1902174

Countersigned External

………………………. ……………………….

Dr. Mohammad Shahed Hasan Khan Tushar

Head Professor
Department of Mechanical Engineering Department of Mechanical Engineering
RUET, Rajshahi-6204 RUET, Rajshahi-6204

iii | P a g e
 Acknowledgements

At first, Authors would like to thank to Almighty for his blessing and support. It is a great pleasure
and opportunity for authors to work and present the report on “Design, Development and
Wireless Control of a Flexible Robotic Arm”. Authors would like to express their sincere
appreciation, deep gratitude, and respect to the supervisor, Professor, Dr. Md. Rokunuzzaman,
Department of Mechanical Engineering, (RUET) for his outstanding supervision, continuous
guidance, inspiration, and support throughout the journey from selecting thesis topic to finalize
the report.

Authors want to express their deepest sense of gratitude and gratefulness to professor Dr.
Mohammad Shahed Hasan Khan Tushar, Head of the Department of Mechanical Engineering,
Rajshahi University of Engineering and Technology.

Authors would like to thank their family members and friends for their peaceful cooperation
throughout the journey of thesis.

June 2025 Authors

RUET, Rajshahi Md. Asifur Rahman Sakib
Samit Sharma

iv | P a g e
 Abstract

This research presents the comprehensive design, development, and wireless control
implementation of a flexible robotic arm featuring four degrees of freedom (4-DOF). The primary
aim is to overcome the inherent limitations of conventional rigid robotic arms, which often struggle
in environments requiring compact maneuverability, adaptability, and safe human interaction. To
achieve these goals, a cost-effective and modular system was developed using 3D-printed PLA
components for the mechanical structure, MG996R servo motors for joint actuation, and a master-
slave wireless control framework based on Arduino microcontrollers and HC-05 Bluetooth
modules. A series of experimental tests involving the manipulation of objects with diverse shapes,
sizes, and weights were done to evaluate the performance of the robotic arm. To assess the system’s
functionality some metrics such as grab and release time, stability, and payload handling were
used. The results indicate that the arm excels at handling flat, medium-sized objects with high
reliability and consistency. Additionally, the wireless control system enabled smooth, real-time
operation without physical tethering, which enhances deployment flexibility. This flexible robotic
arm prototype offers a promising solution for applications in remote handling, field inspection,
light industrial automation, and assistive tasks in healthcare settings. Its modular architecture, low
production cost, and reliable wireless operation make it an accessible and adaptable platform for
further research and practical use in environments where traditional robotic arms may be
unsuitable.

v|Page
 List of contents

Chapter Title Page No.

Certificate iii
Acknowledgement iv
Abstract v
List of contents vi-viii
List of figures ix
Abbreviations and list of symbols x
List of tables xi

1 Introduction 01-06

1.1 Background
1.2 Statement of the Problem
1.3 Motivation
1.4 Research Gap
1.5 Objectives
1.6 Scope and Limitations

2 Literature Review 07-13

2.1 Overview
2.2 Literature Review of Existing Work

3 Methodology 14-36

3.1 Overview
3.2 Methods and Tools used

vi | P a g e
 3.3 Design of the Experiment set up
3.3.1 Calculations
3.4 3D printing of the Prototype
3.5 Experimental Setup
3.5.1 Main Components of the setup
3.5.1.1 Arduino Uno R3
3.5.1.2 Arduino Pro Mini
3.5.1.3 HC-05 Bluetooth Module
3.5.1.4 Relay Module
3.5.1.5 Buck Converter
3.5.1.6 Perfboard
3.5.1.7 Servo Motors
3.5.1.8 Rechargeable Battery
3.5.1.9 Payloads
3.6 Controlling System Design
3.6.1 Circuit Design
3.6.1.1 Drive Power Circuit
3.6.1.2 Motion Power Circuit
3.7 Cost Estimation of the Project

4 Result and Discussion 37-41

4.1 Overview
4.2 Performance Test
4.3 Degree of Freedom
4.4 Discussion

5 Conclusions and Future Recommendations 42-43

5.1 Conclusions
5.2 Future recommendations

vii | P a g e
References 44-47

Appendix 48-54

viii | P a g e
 List of figures

Figure No. Title Page No.

1.1 Annual Publication Trends from 2016 to 2025 02
3.1 Research and Development Process Flow 14
3.2 Proposed System Architecture of a Flexible Robotic Arm 15
3.3 The drawing of components of the prototype 18
3.4 Assembly of components of the prototype 19
3.5 3D printing and the assembly of the prototype 21
3.6 Experimental setup for Grabbing and releasing different objects with 22-23
variety of shapes
3.7 Pin Diagram of Arduino UNO. 25
3.8 Arduino Pro Mini 26
3.9 HC-05 Bluetooth Module Circuit Connections 27
3.10 5V Single-Channel Relay Module 28
3.11 Buck Converter Basics 29
3.12 Copper Perfboard 29
3.13 MG90S Metal Gear Servo Motor 30
3.14 Rechargeable Battery (3.7V×4) 31
3.15 Payloads with variable shapes 32
3.16 Drive power circuit 34
3.17 Theoretical Motion Power Circuit 35
3.18 Practical Motion Power Circuit 35
4.1 Grab and release times for various objects 38
4.2 Kinematic Diagram of a Multi-Link Robotic Arm 39

D1 Similarity report
51

ix | P a g e
 Abbreviations and list of symbols

Abbreviations:

4-DOF: Four Degrees of Freedom

PLA: Polylactic Acid
MG996R: Model name of servo motor
HC-05: Bluetooth module model
PWM: Pulse Width Modulation
IoT: Internet of Things
PVT: Photovoltaic Thermal
NEPCM: Nano-Enhanced Phase Change Material
URLLC: Ultra-Reliable Low-Latency Communication
MPC: Model Predictive Control
HRI: Human-Robot Interaction

Symbols:
L₁, L₂, L₃, L₄: Lengths of robotic arm segments
m₁, m₂, m₃, m₄: Masses of robotic arm segments
g: Acceleration due to gravity (9.81 m/s²)
τ: Torque
α: Angular acceleration
I: Moment of inertia
mp: Payload mass
d: Distance from joint

Subscripts
elbow: Refers to the elbow joint of the robotic arm
dynamic: Related to dynamic load
total: Sum or combined value
payload: Related to the payload being handled

x|Page
 List of tables
Table Title Page
No. No.
3.1 Methods and tools used 15-16
3.2 Cost analysis of the project 36
4.1 Grasping and Releasing Times for Various Object Shapes with 37-38
Corresponding Physical Properties.

4.2 Degree of freedom of the prototype 40

B1 Hardware Components 49
B2 Software/Tools 50

xi | P a g e
 Chapter 1
Introduction
1.1 Background
The evolution of robotic arms is a prime example of one of the most significant achievements in
engineering in modern manufacturing paradigms. Starting in the 1960s with the addition of the
Unimate at General Motors [1], mechanical arms have undergone dramatic changes in both design
ethos and functional capacities. Initial industrial uses were all hydraulically based actuation
systems, which, while rugged, were restricted to around three degrees of freedom with
repeatability specs hardly better than ±1.5mm [2]. This inherent restriction pretty much confined
their use to straightforward material handling operations in highly regulated environments.
There came a dramatic change in the 1980s with the use of commercial electric servo motors
coupled with harmonic drive reducers. Research by the Tokyo Institute of Technology showed that
this setup managed to provide positional accuracy within 0.1mm, with less than 2-second cycle
times for operation [3]. The technological advances enabled extensive use of six-degree-of-
freedom articulated arms in car assembly lines, especially in precision welding, where thermal
distortion issues had earlier posed serious quality challenges [4].

Figure 1.1: Annual Publication Trends from 2016 to 2025

Figure 1.1 illustrates a steady rise in total publications from 2016 to 2023, peaking at 27,945. A
notable decline is observed in 2025, indicating a potential shift or disruption in research output.

1|Page
The medical robotics revolution in the late 1990s brought about new paradigms in design that
emphasized flexibility in rigid-link mechanisms. The da Vinci Surgical System's proprietary
"wristed" devices [5] showed the kind of precise mechanical engineering that could provide seven
degrees of freedom and yet maintain the rigidity required to provide sub-millimeter accuracy. Later
clinical studies showed that these systems enabled patient recovery times to be decreased by
approximately 40% when compared to traditional laparoscopic techniques [6]. Parallel advances
in industrial robotics produced collaborative robot (cobot) designs with torque sensors that were
able to sense human contact within 5ms, making it possible for safe operation without safety cages
[7].

Current challenges involve removing control cabling without sacrificing performance

requirements. Recent experiments using 5G URLLC (Ultra-Reliable Low-Latency
Communication) networks have demonstrated closed-loop control latencies of 2.8±0.3ms under
laboratory conditions simulated [8]. However, real-world manufacturing environment field tests
did reveal occasional spikes in latency to 8-12ms at peak network loading [9]. Materials
technology has also progressed in parallel, with composite carbon fiber arms now having vibration
damping coefficients 30% higher than equivalent aluminum models but weighing 45% less [10].
These evolutions pose new challenges, notably in terms of thermal stability; the thermal expansion
coefficient of such composites can differ by as much as 15% in different axes, which necessitates
the use of advanced compensation algorithms [11].

The advent of machine learning methods has opened up new possibilities for solving long-standing
control issues. Reinforcement learning methods have been shown to be very successful in
managing the kinematic redundancy of hyper-redundant manipulators, achieving a 60% decrease
in path planning computation times compared to conventional methods [12]. These methods are,
nonetheless, computationally demanding, with present implementations needing GPU-accelerated
hardware to ensure continued real-time operation readiness [13]. Energy efficiency is a key field,
since demands for unrestricted performance consistently eclipse advances in battery technology.
The newest lithium polymer cells currently available still restrict continuous operating times to
approximately 2.5 hours for standard loads [14].

2|Page
1.2 Statement of the problem

Due to the growing complexity of industrial automation, medical procedures, and dangerous
environment operations, conventional rigid robotic arms are severely constrained in tight spaces
and delicate tasks. Their rigid bodies, wired control systems, and low flexibility easily result in
poor maneuverability, human collaboration safety problems, and inefficiency in dynamic
environments. As a result, industries incur tremendous losses in productivity due to equipment
downtime, while medical and rescue applications experience degraded accuracy in critical
procedures.

In order to overcome the challenges above, we suggest creating and designing a wirelessly
controlled flexible robotic arm system that will provide rigid-link accuracy with greater
maneuverability. The proposed solution will have advanced materials for improved strength-to-
weight ratio, low-delay wireless control for complete freedom of motion. The system will exhibit
exceptional dexterity in tight spaces, better human-robot collaboration with fewer safety hazards,
and reliable performance in harsh conditions where conventional robotic arms are restricted.

1.3 Motivation

The motivation behind developing a flexible robotic arm stems from the growing demand for
adaptable, efficient, and safe robotic solutions across various fields. Traditional rigid robotic arms
often suffer from limited range of motion and poor adaptability, making them less effective in
performing complex tasks, especially in dynamic or confined environments. In contrast, a robotic
arm with flexible joints offers enhanced maneuverability without compromising strength or
durability, allowing it to execute more intricate operations. The design of such arms using hard but
lightweight materials also enables simplified mechanical systems—requiring fewer motors while
achieving a wide range of movements. This not only optimizes system performance but also
reduces overall cost and complexity.

Moreover, integrating wireless control into the system brings significant advantages. It eliminates
the constraints of wired connections, allowing for untethered, clutter-free operation with greater
mobility. This makes the system suitable for remote deployment, which is essential in modern

3|Page
applications such as field robotics, remote inspection, and disaster response. A powerful yet
flexible robotic arm can also function effectively in environments where conventional arms are
either too large or incapable of navigating, such as cramped repair zones or mobile robotic
platforms.

Additionally, in human-robot interaction (HRI) scenarios, safety and ergonomics are paramount.
Flexible arms made from strong but safe materials reduce the risk of injury during close
collaboration with humans, making them ideal for use in medical assistive technologies and
cooperative manufacturing lines. Finally, the portability and adaptability of such systems allow
them to be mounted on drones, rovers, or other mobile platforms, enabling a wide range of outdoor
and off-site operations in agriculture, defense, mining, and beyond.

1.4 Research Gap

Despite growing interest in flexible and wireless robotic systems, several key gaps remain
unaddressed. First, there is limited integration of flexibility and wireless control in low-cost
platforms like Arduino, with most commercially available systems being either expensive or
tethered. Secondly, real-time Bluetooth-based control for multi-DOF robotic arms remains
underdeveloped, especially for smooth, joint-specific actuation. Third, many existing low-cost
designs lack evaluation of critical performance metrics such as payload capacity, repeatability, and
object adaptability. Additionally, intuitive human-robot interaction interfaces are scarce, often
requiring complex coding and reducing accessibility for non-experts. Finally, current flexible arm
designs offer limited modularity and scalability, hindering system upgrades or extensions without
a complete redesign.

1.5 Objectives

This research aims to design, implement, and evaluate a low-cost, flexible robotic arm system
with wireless control. The key objectives are:

1. Design and Fabrication of a Flexible Multi-Joint Arm:

To conceptualize and build a lightweight, modular robotic arm with multiple degrees of

4|Page
 freedom, using suitable materials and joint types to enable smooth, coordinated motion.
The design will ensure adaptability and support customization for tasks such as pick-and-
place and object manipulation.

2. Wireless Control Integration:

To develop a Bluetooth-based wireless control system using Arduino hardware that enables
real-time, stable communication. A user-friendly interface will allow remote operation
with minimal delay, enhancing human-robot interaction.

3. Performance Evaluation:

To experimentally assess the arm’s payload capacity, positional repeatability, and object
handling capability. Tests will simulate real-world use cases, providing quantitative data
to validate the system’s effectiveness in applications like light automation, education, and
assistive tasks.

1.6 Scope and Limitations

The project involves the development of a flexible robotic arm featuring modular joints created
through 3D printing technology. This modular approach allows for easier assembly, maintenance,
and potential future modifications. The robotic arm is controlled wirelessly using HC-05 Bluetooth
modules integrated with an Arduino-based architecture, enabling remote operation without
physical tethering. The system is tested through a series of object handling trials aimed at
evaluating the arm’s adaptability to different shapes, its response time, and the stability of grasp
and movement during manipulation. Potential applications for this flexible robotic arm include
educational demonstrations to showcase robotics principles, light industrial tasks that require
handling of small or delicate objects, remote manipulation in environments where direct human
presence is difficult or unsafe, and assistive robotics to aid individuals in daily tasks.
However, the system has several limitations. The control of the robotic arm is not autonomous and
relies entirely on manual input through Bluetooth communication, which limits its ability to
perform tasks independently. Environmental testing was confined to standard laboratory
conditions, and performance under extreme thermal conditions outside the range of 0 to 50 °C or
in high-radiation environments was not evaluated. Additionally, the design focuses on flexible
robotic arms constructed from rigid-link modular joints and does not explore continuum or soft-

5|Page
bodied robotic models, which might offer greater flexibility but involve more complex design and
control challenges. These limitations define the current boundaries of the project and provide
directions for future enhancements.

6|Page
 2 Chapter 2
Literature Review
2.1 Overview
The literature highlights advancements in cable-driven and flexible robotic arms, focusing on
design, control, and adaptability. Key developments include lightweight structures, adaptive
stiffness, and precise control systems. These innovations enhance performance in surgery, space,
and remote operations, supporting the development of high-precision flexible arms.

2.2 Literature Review of Existing Work

The article "D3-ARM: High-Dynamic, Dexterous, and Fully Decoupled Cable-Driven Robotic
Arm," by Hong Luo, Jianle Xu, Shoujie Li, Huayue Liang, Yanbo Chen, Chongkun Xia, and
Xueqian Wang (2025), presents a new cable-driven robotic arm that overcomes frequent problems,
including motion coupling and cable slack. Through the use of a low-friction motion decoupling
mechanism at every joint, as well as the addition of a cable pretension system, the D3-ARM is
capable of high precision and stability. The robotic arm with six degrees of freedom measures 776
mm in length and has a total weight of only 1.6 kg. Experimental findings exhibit a mean position
error of 1.29 mm and a payload capacity of 2.0 kg, thereby showing the efficacy of the proposed
design to improve control precision and performance overall [15].
Burgner-Kahrs, Rucker, and Choset published a paper in 2015 on a type of robot known as a
continuum robot. It does not have stiff joints as a normal robot arm does. Instead, it curves like a
snake or an elephant trunk, bending smoothly from end to end. Its adaptability in these matters is
what renders it useful in narrow or irregular places, such as inside a human body in surgery. They
describe how the robot function, how they are controlled, and some issues with making them, as
well as using them. Issues include obtaining accurate feedback or making them move as required.
They also explain how such robots are more flexible and versatile than normal robots. This article
is useful for our research as it discusses a lot of fundamental concepts involved in making flexible
robotic arms, particularly movement and control [16].
In their 2021 study, Zhou, Lin, and Wang worked on building a robotic arm that can change how
stiff or flexible it is while it's operating. They used a cable-driven design to keep it lightweight,
which also helped with smooth movements. The main idea was to let the arm become soft or stiff

7|Page
depending on the task, especially when working near people or delicate objects. The paper talks
about how they made the stiffness control work and how it helped during testing, like when the
arm had to deal with different types of contact. For our thesis, this paper gave us ideas on how to
improve flexibility and control in robotic arms, especially by using smart cable systems and
adaptable stiffness [17].
Berkelman and Ma created a small robotic arm in 2016 with two types of movement via cables.
The arm is very useful in confined spaces but still has a decent range, making it convenient for
small-space work. They made an effort to organize the cables and pulleys in a way such that
movement would be smooth yet accurate with a minimalist design with minimal weight. In this
paper, they detail how they constructed parts of the system and how they worked in experiments.
For my thesis, this paper provided me with concepts on how devices using cables are constructed
with a balance between size, flexibility, and accurate movement, which is what we are seeking
[18].
Tang, Liu, Zhao, Wang, and Zhang published an article in 2020 on a robotic arm using cables.
They made a light, space-suitable arm, where being light and trustworthy is crucial. They wanted
a strong arm but one with some flexibility for performing delicate tasks in a zero-weight
environment. They explained how they structured the cables and joints in a way that would
enhance motion as well as control without increasing the arm's weight. They also tested the arm's
capabilities under space-like conditions in order to demonstrate how it would perform with regular
tasks inside an orbital environment. We found this research useful for our thesis because it presents
some great concepts for developing accurate, light cable-driven arms for challenging environments
[19].
Yang, Li, and Wang made cable-driven flexible robot arms in 2022. They discovered that it was
difficult to precisely control the arm when cables bent and stretched. They then constructed a
detailed model in order to comprehend how the cables and arm collaborated. They then formed a
control system using this model, one that would address these issues and continue having the arm
follow precisely to the right locations. They implemented their plan with great results, with it
improving how precisely the arm could go to specific locations. In our thesis, this paper is
extremely useful as it provides concepts on modeling as well as controlling flexible cable-driven
arms at high accuracy and performance [20].
This 2016 paper by Murray examined a cable-based flexible surgical robot designed for precise

8|Page
movements in confined spaces. As opposed to normal stiff surgery tools, this robot can navigate
small and curved paths and is excellent for minimally invasive surgery. The paper discussed how
the cable system provides the robot with high levels of flexibility and stability. Murray described
how the cables were arranged to prevent them from being loose and to provide precise force
transmission in a very crucial task such as surgery. Through the tests, it was apparent from the
results that the robot maintain control while moving over complicated paths. This work is relevant
to our thesis because it presents a design with high precision and flexibility desired for high-rate
movement and stability [21].
Xu, Y. Li, and H. Liu published a paper in 2018 explaining a unique robot designed to be used in
minimally invasive surgery. The purpose was to design a flexible robot capable of moving through
complex and narrow passages in the body and remain precise. The design consisted of a number
of cables running down a flexible backbone and attaching to actuators. The authors also
implemented a combination of position and force control techniques, which enabled the robot to
perform suitably with soft tissues and remain stable. The robot was tested on a model and
demonstrated being able to perform precise and reproducible work. This paper is relevant to our
thesis because it lays out how cable-actuated arms can be designed to be precise and rapid enough
for fine work in cramped spaces [22].
J. Chen, M. Zhao, and L. Zhou introduced a landmark paper in 2020 exploring the design of a
modular cable-driven flexible manipulator. They proposed a system where segments of the arm
were interchangeable or reconfigurable as per the task's requirements, thus imparting an incredible
level of versatility to the system. Each segment was designed by combining cables and flexible
joints to allow for effortless bending and fine-grained control over movement. The design was
modular and thus made maintenance simple but also provided scalability by allowing the
manipulator to be adapted for a variety of different environments and applications. Additionally,
a new control method was proposed by them where movements of different segments were
synchronized in a smooth and stable manner. This work resonates with our thesis because it shows
how modularity in cable systems can provide high levels of versatility without compromising on
flexibility and accuracy [23].
Kim and Park published a paper in 2018 detailing how to make flexible robotic arms even more
accurate by adjusting cable tension. They determined that small errors in tension resulted in a lot
of shake or wobble, particularly in light or flexible arms. Therefore, they developed a control

9|Page
system that adjusts tension in real-time, making the arm move with greater accuracy and be able
to adjust to changing conditions. They experimented with this with a model and demonstrated that
with their approach, errors decreased and movement was smoother, even with varying weights.
This is beneficial to our thesis as it presents sound ideas for enhancing the accuracy and stability
of flexible arms through smart tension control [24].
Torres and Murphy published a paper in 2019 on cable-driven robotic arms. They described how
to design and control the robots to move with precision and in various manners, similar to a human
limb. They mentioned problems with cable movement, tension, and ensuring that the robots move
reliably and smoothly. They also described the need for a balance between strength and flexibility,
something that is necessary for robots in sensitive or cramped environments. This paper is crucial
for our thesis as it contains valuable suggestions for the construction of flexible as well as easy-
controllable cable-driven robotic arms [25].
In a paper published in 2019, Li, Xu, and Hu discussed a flexible manipulator which was designed
to be used for remote operations. The manipulator was designed to have flexibility to bend and
flex for hard to reach areas still providing the manipulator having strength and accuracy. The paper
explains how the manipulator's structure and cable system work together to provide flexibility
without reducing control accuracy. Also, the authors investigated control strategies to stabilize the
arm moving through complex or remote operations. Stabilizing the arm is essential during
operations remote from the user or in dangerous environments. Overall, this research will be
valuable for our thesis because it provides practical design and control ideas for flexible cable-
driven arms designed for sensitive remote operation [26].
In 2020, Savastano, Ruggiero, and Siciliano published a paper on modelling and control of a cable-
driven rigid-link manipulator. They were in the process of developing robotic arms which use
cables to location actuator driven rigid segments which couples the rigidity of rigid links and the
versatility of cable actuation. Their research contains a lot of new mathematical models that can
provide an insightful and accurate prediction of the manipulator behaviour, and their efforts to
design control structures for the manipulator that will allow it to complete precise and stable
movements. This research is helpful to our thesis as it gives a good foundation approach to
controlling cable driven rigid-link arms, this will give me better accuracy and flexibility while
applying novel motion techniques to those types of robotic systems [27].
Wang, Hu, and Cao (2021) designed a rigid cable-driven manipulator specifically for assembly

10 | P a g e
tasks. They intended to combine the precise application and rigidness of assembly tasks, along
with lightweight and flexibility afforded by cable-driven actuation. Wang, Hu, and Cao focused
their paper on design, kinematics, and control. Their goal was for accurately positioning and stable
operation in noticeably complex assembly tasks with a cable-driven manipulator. To validate the
research, they showed that the manipulator was capable in realistic assembly tasks. Their research
is valuable to our thesis as it provides a practical way to use cable-driven rigid manipulators in
precision industrial tasks [28].
Sun, Xu, and Wang (2019) designed a feedback control system for a flexible robotic arm to
improve control precision and stability. The feedback control system utilized sensor data that was
transformed to detect changes in the robotic arm's position to continuously adjust the movement
of the robotic arm to compensate for any flexibility-induced deviations. Their controller designs
were also put through rigorous testing to accommodate for the nonlinear behaviors of flexible
arms, in addition to improving general responsiveness and accuracy. The data from both trials
illustrated that their feedback control system was able to reduce the tracking errors as well as
establish better robustness to disturbances. This paper is of great worth to our thesis as it provides
practical methods for increasing the control accuracy of flexible cable-driven robotic arms [29].
The study conducted by Liu and Chen (2020) on adaptive control strategies for flexible cable-
driven robot arms primarily concerned the development of a controller for such systems that could
automatically compensate for changes in the arm due to both flexibility and external disturbances.
Their control strategy implemented an adaptive control method that estimated the unknown
parameters in real time. The effects of the unknown parameters were minimized, and the stability
and accuracy of the robot arm's tracking capabilities were improved over a variety of tasks.
Experimental data from their research showed that their experimental adaptive controller was able
to track error more closely than fixed-parameter controllers, and provided additional robustness.
This research is valuable for our thesis as it showed a viable method of dealing with uncertainties
of the flexible cable-driven robot arm [30].
Xu, Lin, and Yu (2021) detailed the design and validation of a lightweight cable-driven robotic
arm with a goal of medical teleoperation. Their system was small, fast, and light weight to provide
safe and accurate manipulation in delicate environments. The robotic arm had a modular design
and integrated high torque cables in a way to eliminate backlash and still allow flexibility. The
control implementation combined position tracking with haptic feedback to give remote operators

11 | P a g e
greater accuracy. Simulation and bench tests showed the implementation had low latitude with
accurate fidelity of the motion translated. This study helps bolster our thesis because it adds to the
evidence that cable driven designs can be adapted in the teleoperated medical field while
maintaining the balance of lightness, precision and safety [31].
Zhao and Li (2022) investigated a hybrid stiffness control method for cable-driven robotic arms in
order to produce a robotic arm that exhibited good flexibility and load-bearing capacity. The mixed
stiffness method combines active cable tension control and passive compliance control to allow
the arm to dynamically control its stiffness based on the demands of the task. The authors created
a control model which included force feedback (cable tension) and joint deformation
measurements (compliance) to help strike the balance between accuracy and safety. Results from
experiments showed that the system could successfully adapt its stiffness in real-time and
maintained a steady equilibrium for all payloads, even when interacting with a human. This is
significant, as it informs our thesis as it demonstrates the potential for adaptive stiffness systems
to be introduced to cable-driven arms to allow for dexterous manipulation and compliant
interaction [32].
Chen, Guo, and Zhang (2022) presented a lightweight cable-driven robotic arm designed for high-
precision tasks in confined spaces. An optimal structural design was adopted to make the structure
as small as possible while maximizing the performance of the actuation devices. A major
innovation was a cable routing arrangement that reduced backlash and cable efficiencies, forcing
efficiencies. A precision tracking control algorithm was defined using real-time joint position
feedback for correcting elastic deformation in the cables. In validation studies it was shown that
the arm could successfully conduct careful operations with accuracy in the millimeter region. This
study supports our thesis, showing how compact structural design, and integration with control
can improve the performance of flexible robotic arms in fine manipulation tasks [33].
Park, Lee, and Kim (2022) sought to enhance the flexibility and response of cable-driven robotic
arms by embedding a distributed sensor feedback system into the cable actuation lines. The authors
used strain gauges, IMUs, and tension sensors in a unique way to capture the live usage behaviors
of cables, real time joint deformation with respect to the cables, and adjustments to accommodate
all manipulative forces on the cables. A sensor fusion algorithm was embedded to correct
positional error in real-time and provide adaptive compensation when dynamically performing
tasks. The authors showed substantial improvements in tracking with respect to positional error in

12 | P a g e
the cases of altered loads on the cables and whenever the cable-driven robotic arm changed its
direction rapidly. This paper supports our thesis by illustrating the applicability of embedded
sensing for precise wireless control and adaptive control of flexible and adaptive mechanical
system [34].
Real-time model predictive control (MPC) for continuum robots was first introduced by Rucker
and Choset (2018) with the goal of ensuring that motion planning and control are accurate in
unstructured environments, thereby focusing on a given task. Rucker and Choset, manipulated a
reduced-order dynamic model to allow for quicker optimization while capturing the required
behavior of the system, into account. In addition, their method is constrained by cable tension,
environmental contact, actuation limits, and the MPC cycle allows then to be both responsive and
stable while their flexible arm was in action. The authors demonstrated simulations to hardware
and developed a trajectory tracking system with real-time controls that enabled an accurate
tracking of a complex trajectory. Rucker and Choset (2018) is useful to our thesis because this
paper provides a control architecture for a continuum robot that supports real-time wireless
actuation specifications for flexible robotic arms [35].

13 | P a g e
 3 Chapter 3
Methodology
3.1 Overview
The primary goal of this project is to design and develop a flexible robotic arm capable of
performing precision tasks through wireless control. This system aims to overcome the limitations
of traditional rigid robotic arms by introducing adaptability in handling various shapes and sizes,
especially in environments with dynamic conditions. By integrating 3D printing technology and
stepper/servo motor-based models, the arm is optimized for applications in fields such as
healthcare, manufacturing, and research, where precision and flexibility are essential.

Figure 3.1: Research and Development Process Flow

Figure 3.1: represents a structured, step-by-step approach starting with a comprehensive literature
review and defining clear objectives. After identifying the core problem, a detailed study and
necessary assumptions/calculations are made to inform the arm design. The finalized design is
then realized using a 3D printer, followed by assembly and control circuit setup. These components

14 | P a g e
are integrated in the system integration phase, leading to performance testing. Finally, results are
analyzed in the discussion phase, concluding the project with key insights.

Figure 3.2: Proposed System Architecture of a Flexible Robotic Arm

Figure 3.2: represents a wireless control system designed for operating a soft robotic arm. On the
transmitter side, user input is provided through a switch, which is read by an Arduino Pro Mini
powered by a 5V battery. This Arduino processes the input and transmits the corresponding signal
via a Bluetooth module. On the receiver side, another Bluetooth module receives the signal and
passes it to an Arduino Uno. The Arduino Uno then activates a relay module, which controls the
servo motor responsible for actuating the robotic arm. The servo motor is powered by a separate
5V battery to ensure stable operation. This system enables effective wireless communication and
actuation, making it suitable for controlling a flexible or soft robotic arm.

3.2 Methods and Tools used

Table 3.1: Methods and tools used

Items Specifications
Code Development Arduino IDE

Development Platform VS code

Language Used C++

15 | P a g e
 Drawing Platform Fusion 360
Manufacturing 3D printing

Material Used PLA

Motor Used Stepper, Servo
Circuit Design Proteus

Wireless Communication HC-05 Bluetooth Module

Microcontrollers Used Arduino Pro Mini, Arduino Uno
Control Method Master-Slave

3.3 Design of the Experimental set up

The experimental setup for the flexible robotic arm was designed to evaluate its mechanical
performance, wireless control capability, and object-handling efficiency. The system consists of
both hardware and software components integrated into a modular framework.

Mechanical Structure
The robotic arm is constructed using lightweight 3D-printed parts to ensure cost-effectiveness and
flexibility. It features:
 A base and stand for structural support
 Four rotational joints, each driven by a servo or stepper motor
 Linkages modeled and analyzed using SolidWorks for load-bearing capability

. Relevant Dimension of the prototype is given below:

 Length of the Prototype

= (Base joint length + Stand joint + Joint 1 + Joint 2 + Joint 3 + Front Joint)
= (5+4.5+10+10+10+8.5) cm
= 48 cm
 Width of the Prototype = 3 cm
 Height of the Stand = 21.5 cm

16 | P a g e
  No. of joints of the arm = 4
 Dimension of the Servo motor

MG996R Servo Motor

 Torque: ~20 kg·cm (at 6V)

 Operating Voltage: 4.8V – 7.2V
 Dimensions: 40.7 mm × 19.7 mm × 42.9 mm
 Weight: ~55 grams
 Stall Torque (6V): ~1.2–1.3 Nm (12–13 kg·cm)
 Rotation Range: 180° (±10°)

Figure 3.3 depicts the finalized drawing of the prototype, considering the specified dimensions.

Figure 3.3 (a): Base Figure 3.3 (b): Stand

17 | P a g e
Figure 3.3 (c): Stand Joint Figure 3.3 (d): Fixed Link

Figure 3.3 (e): Link 1 Figure 3.3 (f): Link 2

Figure 3.3 (g): Link 3 Figure 3.3 (h): Front Link

Figure 3.3: The drawing of components of the prototype

18 | P a g e
 Figure 3.4: Assembly of components of the prototype

Figure 3.4 shows a 3D-rendered view of the fully assembled prototype robotic arm, highlighting
the arrangement and integration of its mechanical components.

3.3.1 Calculations

Static Load (Torque) Calculation

Arm links: L1 = L2 = 0.15 m, m1 = m2 = 0.1 kg.

Gravity: g = 9.81 m/s2.

Elbow Joint:
τelbow = m1g(L1/2) + m2g(L1+ L2/2) + m3g(L1+ L2+ L3/2) + m4g(L1+ L2+ L3+ L4/2) + + m5g(L1+
L2+ L3+ L4+L5/2)
= 0.012+0.032+0.+0.142+0.24+0.338+0.429
=1.193 Nm

Dynamic Load (Torque) Calculation

Angular acceleration: α = 1rad/s

1
Moment of inertia for a rod about its center: Irod = 12m𝐿2
Moment of inertia about some distance: I = (1/12) mL2+m(distance from joint)2

Elbow Joint
1 1 1
= 12 (0.1)(0.1)2+(0.1)(0.1425)2 + 12
(0.1)(0.1)2+(0.1)(0.0425)2 + 12 (0.1)(0.1)2+(0.1)(0.0575)2 +
1
(0.085)(0.085)2+(0.085)(0.15)2
12

19 | P a g e
Total Inertia at Elbow

Itotal,elbow = 0.0046 kg·m2

Dynamic Torque at Elbow:

τdynamic,elbow = Itotal · α = 0.0046 · 1 = 0.0046 Nm.

Total Torque at Elbow:

τtotal,elbow = 1.193+0.0046 = 1.197 Nm.

Maximum Payload Calculation

The payload adds static and dynamic torque at the elbow:

τelbow, payload = mpgL+Ipayload, elbow *α
= mp*9.81*0.24 + = mp*(0.24)2*1
= 2.41 mp Nm.

Constraint:

1.19 + 2.41 mp ≤ 2.0825

2.41 mp ≤ 0.8925
0.8925
mp ≤ » 0.375kg.
2.41

The 1st joint is the limiting factor. The maximum payload the arm can carry is 0.375 kg (375
grams) while staying within the torque limits and safety factor.

3.4 3D Printing of the Prototype

The mechanical framework of the developed robotic arm is entirely constructed using additive
manufacturing (3D printing), allowing for a lightweight, modular, and cost-efficient design. Figure
3.5 shows the structure comprises a circular base platform and a vertical stand that supports the
entire arm assembly. These foundational components ensure stability during movement and load

20 | P a g e
handling.
The robotic arm consists of multiple interlinked joint modules, each providing rotational motion
about a single axis. The servo motors are directly embedded at these joints to enable actuation, and
the printed enclosures are tailored to house both the motors and internal wiring. This design not
only minimizes external clutter but also enhances mechanical protection and simplifies assembly.

Figure 3.5 (a) Figure 3.5 (b)

Figure 3.5 (c)

Figure 3.5: 3D printing and the assembly of the prototype

21 | P a g e
Each arm segment is fabricated as a hollow structure, allowing internal routing of signal and power
wires from the base to the end-effector. The use of servo motors at each joint provides discrete
angular control, collectively enabling four degrees of freedom in the X-Y plane. The modular
design of the segments further permits easy customization or replacement of parts.
At the distal end of the arm, a 3D-printed gripper mechanism (end-effector) is integrated, capable
of grasping and manipulating objects with varying shapes and sizes. The functionality of the end-
effector is demonstrated in experiments where it successfully grips test objects such as cylindrical
and rectangular items, confirming the precision and mechanical repeatability of the system.
This design showcases how low-cost fabrication techniques, combined with embedded actuation
and integrated wiring, can produce a functional, adaptable, and wireless robotic arm suitable for
diverse applications such as object handling and educational demonstrations.

3.5 Experimental Setup

Figure 3.6 (a–g) illustrates the functional performance and workspace environment of the
developed robotic arm. In Figure 3.6(a), the robotic arm grips a small green plastic basket,
demonstrating its ability to handle lightweight, hollow objects.

Figure 3.6(a) Figure 3.6(b)

22 | P a g e
Figure 3.6(c) Figure 3.6(d)

Figure 3.6(e) Figure 3.6(f)

Figure 3.6(g)

23 | P a g e
Figure 3.6: Experimental setup for Grabbing and releasing different objects with variety of shapes.

Figure 3.6(b) shows it holding a plastic bottle with a red cap, highlighting its adaptability to various
object shapes. Figures 3.6(c) and 3.6(d) capture the arm in different motion states while gripping
or releasing objects, showcasing its degrees of freedom and range of motion. Figures 3.6(e)
presents the control board—likely an Arduino or similar microcontroller—connected with several
wires, indicating the real-time embedded control system used to operate the arm. The setup view
in Figure 3.6(f) shows the arm in an idle or preparatory position, surrounded by objects like books,
offering insight into its reach and stability in a semi-cluttered environment. Lastly, Figure 3.7 (g)
captures the arm gripping two different objects simultaneously, demonstrating its potential for
sequential task execution or dual-object recognition. Altogether, the figure highlights the robotic
arm’s mechanical capabilities, control integration, and adaptability in real-world manipulation
tasks.

3.5.1 Main Components of the setup

The project uses various components to implement the system prototype. Inexpensive and
precise components are selected for the system in such a way that it is economical and available
without compromising the quality of the components. The list of components is as follows:
1. Mechanical System (Flexible Robotic Arm)
2. Arduino Uno R3
3. Arduino Pro-mini
4. HC-05 Bluetooth Module
5. Relay Module
6. Buck Converter
7. Perfboard
8. Servo Motors
9. Power Supply/Battery
10. Payloads
11. Stop watch
12. Scale

24 | P a g e
3.5.1.1 Arduino Uno R3

The Arduino Uno R3 is a popular microcontroller board designed for both beginners
and professionals. It is part of the Arduino family of boards and has become one of the most widely
used development platforms due to its simplicity, versatility and affordability.
The Arduino Uno R3 is based on the ATmega328P microcontroller at the heart of the board
[36]. It has 14 digital I/O pins, 6 analog inputs, a 16 MHz crystal oscillator, a USB interface for
programming and power supply, an ICSP header, and a reset button [37].
These features make it easy to connect various sensors, actuators and other electronic components,
allowing users to create a variety of projects. The simplicity of the Arduino Uno R3 is one of its
main benefits. Even for people who have little to no prior knowledge with electronics or
programming, the board is made to be user-friendly. Based on C/C++, the Arduino programming
language is simple to learn and enables users to get started creating and playing with various
projects right away [38]. Writing, compiling, and uploading code to the board is made simpler by
the integrated development environment (IDE) offered by the Arduino software [39]. Another
noteworthy aspect of the Arduino Uno R3 is its adaptability. It can be applied to a wide range of
tasks, from straightforward LED blinking projects to more intricate Internet of Things (IoT),
robotics, and home automation projects [40].

25 | P a g e
 Figure 3.7: Pin Diagram of Arduino UNO [41]

3.5.1.2 Arduino Pro Mini

The Arduino Pro Mini is a compact, low-cost, and flexible microcontroller board based on the
ATmega328 microchip, designed for semi-permanent and space-constrained embedded
applications. It comes in two voltage versions—3.3V (8 MHz) and 5V (16 MHz)—and features
14 digital I/O pins, 6 analog inputs, and support for PWM output. Unlike other Arduino boards, it
lacks built-in USB connectivity, requiring an FTDI programmer for uploading code. Its small form
factor, low power consumption, and affordability make it ideal for wearable electronics, robotics,
and portable devices such as smartwatches, digital thermometers, and line-following robots [42].

Figure 3.8: Arduino Pro Mini

3.5.1.3 HC-05 Bluetooth Module

The HC-05 is a popular and affordable Bluetooth module that enables wireless serial
communication between devices, especially microcontrollers like Arduino. It supports both Master
and Slave modes and communicates via the USART protocol with a default baud rate of 9600.
The module operates in Data Mode for regular communication and AT Command Mode to change

26 | P a g e
settings like name, password, or role.
Powered by +5V, it has a range of up to 100 meters and is ideal for applications such as home
automation, wireless robots, and data logging. It can connect easily with smartphones or laptops
using default credentials (Name: "HC-05", Password: 1234 or 0000). Compact, reliable, and easy
to integrate, the HC-05 is widely used in IoT and embedded systems projects [43].

Figure 3.9: HC-05 Bluetooth Module

3.5.1.4 Relay Module

The 5V single-channel relay module is an electromechanical switching device designed to control

high-voltage or high-current loads using low-power control signals from microcontrollers like
Arduino. It integrates additional components such as transistors, diodes, indicator LEDs, and screw
terminals, enhancing its usability and protection. The relay can handle up to 250VAC or 30VDC
at 10A, making it ideal for mains switching, home automation, and isolated power delivery
applications. It ensures galvanic isolation between the control and load sides, offering both safety
and reliability in embedded systems [44].

27 | P a g e
 Figure 3.10: Relay Module

3.5.1.5 Buck Converter

A buck converter, or step-down converter, is a high-efficiency DC-DC power supply circuit that
reduces input voltage to a lower, regulated output voltage. It operates using key components such
as a switch (MOSFET), inductor, diode, capacitor, and a controller IC, all working together under
PWM control to maintain voltage regulation with minimal loss. Buck converters offer significant
efficiency benefits over linear regulators, particularly in battery-powered and heat-sensitive
applications. They are designed to operate in continuous or discontinuous conduction modes
depending on the load, with synchronous variants offering even higher efficiency by replacing the
diode with a controlled low-side MOSFET. Proper design involves selecting appropriate
component values based on switching frequency, ripple current, and output requirements to ensure
performance, stability, and safety [45].

28 | P a g e
 Figure 3.11: Buck Converter

3.5.1.6 Perfboard

A perfboard (perforated board) is a widely used prototyping platform for building semi-permanent
electronic circuits. It consists of a grid of holes with copper pads where electronic components can
be soldered, offering a more durable and compact solution compared to a breadboard. In this
project, the perfboard was used to ensure stronger mechanical support and stable electrical

Figure 3.12: Perfboard

29 | P a g e
connections for the final hardware implementation. It allows flexible component placement and
manual wiring, making it suitable for low-volume, custom circuit assembly.

3.5.1.7 Servo Motors

The Tower Pro MG90S Mini Digital Servo is a 180° rotation digital servo motor designed for fast
and accurate PWM signal processing. Its advanced internal circuitry ensures reliable torque, strong
holding power, and quick response to external forces, making it ideal for applications requiring
precise control.
Known for its consistent performance and durability, this servo is a preferred choice among RC
enthusiasts. It features a compact and robust plastic housing that provides resistance to dust and
water—an essential advantage for use in RC planes, boats, and monster trucks. The servo comes
with a 3-wire JR connector, which is also compatible with Futaba systems [46].
Wire Configuration:
 Red – Power Supply (Positive)
 Brown – Ground (Negative)
 Orange – Control Signal

Figure 3.13: Servo Motors

30 | P a g e
3.5.1.8 Rechargeable Battery

The 18650 rechargeable Li-ion battery (Button Top, 67×18×18 mm) was used in this project as a
reliable power source. Known for its high capacity, rechargeability (up to 1200 cycles), and no
memory effect, it ensures long-term usability. The battery includes six-layer protection features
such as overcharge, overcurrent, and short-circuit protection, enhancing operational safety. Its
wide compatibility with various portable and electronic devices makes it an ideal choice for
embedded and mobile applications.

Figure 3.14: Rechargeable Battery (3.7V*4)

3.5.1.9 Payloads

Figure 3.15 showcases a diverse set of objects used to evaluate the grasping performance of the flexible
robotic arm, each selected to test specific parameters such as grip adaptability, stability, surface interaction,
and precision. Figure 3.15(a) presents a perfume box, a rectangular prism with a smooth surface and light-
to-moderate weight, ideal for testing how well the arm can grasp flat-edged, medium-sized items and
maintain stability during transport. Figure 3.15(b) features a rectangular-shaped container with a green lid,
filled with small particles. This object introduces cylindrical curvature and moderate to heavy weight,
providing a good test for grip force control and lifting capacity, particularly when grasping around the full
circumference. In Figure 3.15(c), a cylindrical-shaped perforated basket is shown, offering a lightweight

31 | P a g e
and irregularly textured surface due to its holes. It helps assess the arm’s ability to delicately grasp objects
without getting its gripper fingers stuck or deforming the item.

Figure 3.15(a) Figure 3.15(b) Figure 3.15(c)

Figure 3.15(f)

Figure 3.15(d) Figure 3.15(e)

Figure 3.15(g)

Figure 3.15: Payloads with variable shapes

32 | P a g e
Figure 3.15(d) includes a cylindrical-shaped bottle with a red cap, typically used for beverages. Its
tapered body and variable weight depending on content make it ideal for examining vertical
stability and the arm’s adaptability to changing diameters. Figure 3.15(e) depicts a spherical-shaped
ball, likely a ball of yarn or similar, which tests the arm’s ability to grip fully curved surfaces
without allowing the object to roll or slip—focusing on control rather than weight. Figure 3.15 (f)
shows a wallet, a flat, flexible item made of soft material. This object is essential for evaluating
the arm’s ability to gently pick up and manipulate low-rigidity objects from flat surfaces without
pushing or deforming them. Finally, Figure 3.15(g) features a calculator, representing a typical
small electronic device. Its flat, smooth plastic surface and moderate weight make it an excellent
sample for testing precision gripping and slippage resistance.

Together, the objects in Figure 1 (1a–1g) provide a comprehensive range of challenges that mimic
real-world grasping tasks. They allow systematic testing of the robotic arm’s capabilities across
variations in shape, texture, weight, and surface interaction—crucial for optimizing both
mechanical design and control algorithms.

3.6 Controlling System Design

3.6.1 Circuit Design

3.6.1.1 Drive Power Circuit

The drive circuit of the flexible robotic arm integrates both control and power systems to ensure
efficient and stable operation. At the core is the Arduino Pro Mini, which processes commands
from a Bluetooth module (HC-05) and generates PWM signals to control the servo motors. Since
the Pro Mini lacks onboard voltage regulation, a buck converter is used to supply it with a stable
5V from an unregulated DC source. A second buck converter powers the servo motors, providing
a separate 5V–6V line to handle their higher current needs without affecting the control circuit.
PWM signals from the Arduino are distributed to the servos through a perfboard, which replaces
a traditional breadboard for better stability and semi-permanent wiring. A relay module is
optionally used to switch motor power on and off, preventing overheating and saving energy. All
motors are attached to 3D-printed arm components, enabling wireless, precise movement. This
setup ensures the robotic arm operates smoothly with isolated power supplies, reliable

33 | P a g e
communication, and minimal electrical interference.

Figure 3.16(a): Theoretical Drive Power Circuit Figure 3.16(b): Actual Drive Power Circuit

Figure 3.16: Drive power circuit

Figure 3.16 shows the theoretical drive power circuit and the actual drive power circuit.

3.6.1.2 Motion Power Circuit

A motion control circuit is designed to regulate the movement of a mechanical system by

processing input data and controlling actuators like motors. In this project, the core of the motion
control system is a microcontroller board (Arduino Pro Mini), which serves as the brain of the
operation. It interprets user commands and generates PWM signals to control servo motors
accordingly.
The system is powered by a battery, and a buck converter is used to regulate voltage to a safe level
for the microcontroller. A key component in this circuit is the HC-05 Bluetooth module, which
enables wireless communication. It receives commands from a mobile device and transfers the
data to the Arduino, which then processes the information and sends the appropriate signals to the
actuators.

34 | P a g e
 Figure 3.17: Theoretical Motion Power Circuit

Additionally, the HC-05 module serves as a communication bridge between the motion control
circuit and the drive power circuit, ensuring synchronized and wireless operation of the robotic
arm. This setup allows real-time control and smooth actuation of the robotic arm across its joints.

Figure 3.18: Practical Motion Power Circuit

35 | P a g e
Figure 3.17: shows the theoretical drive power circuit and Figure 3.18: shows the actual drive power
circuit.

3.7 Cost Estimation of the Project

Table 3.2 shows the cost analysis of the project in various aspects.

Table 3.2: Cost analysis of the project

Sl. No Item Name Specification No. of Unit Unit cost Total Cost
(Tk) (Tk)

1. Microcontroller Arduino Uno 1 550 550

Board R3

2. Microcontroller Arduino Pro 1 400 400

Board Mini

3. Bluetooth HC-05 2 400 800

Module
4. Vero Board Perfboard 1 750 750

5. Servo Motor MG996R (180 4 475 1900

Degree)

6. Servo Horn - 4 120 480

7. 3.7V Battery Li-ion 5 100 500

8. Buck Converter LM259C 3 100 300

9. PLA material - - - 3850

10. TPU material - - - 2400

11. Case Board 3 200 600

12. Others - - - 500

Total 13,330 Tk

36 | P a g e
 4 Chapter 4
Results and Discussion
4.1 Overview
The performance of the developed flexible robotic arm was evaluated through a series of object-
handling experiments to assess its gripping ability, response time, and payload capacity. The tests
were conducted on objects of varying shapes, sizes, and weights, including rectangular, circular,
and cylindrical forms, in order to simulate real-world grasping tasks and evaluate the arm’s
adaptability.

4.2 Performance Test

During the experimental trials, the flexible robotic arm was tested on a variety of objects differing
in shape, perimeter, and weight to evaluate its grasping and releasing capabilities. As shown in
Table 4.1, the objects were categorized into six shape types: rectangular, cylindrical, spherical,
wallet-like, calculator, and tester. Rectangular objects, including a perfume box and a container,
demonstrated moderate handling times, with the heavier perfume box (347 gm) requiring 5
seconds to grab and 6 seconds to release, while the lighter container (255 gm) required less time
for both actions. Among the cylindrical items, the perforated basket, despite having the largest
perimeter (28 cm), showed efficient performance (6 s grab, 6.5 s release), whereas the bottle, being
lighter (157 gm), surprisingly took longer (8.5 s and 9.1 s respectively), indicating shape-related
gripping difficulties. The spherical ball, although weighing only 70 gm, also required extended
handling times (8.4 s to grab, 8.58 s to release), likely due to its tendency to roll and its curved
surface.

Table 4.1: Grasping and Releasing Times for Various Object Shapes with Corresponding Physical
Properties.

Shape Object Perimeter Grab Time Release Time Weight

Rectangular Perfume 22cm 5s 6s 347gm

box

Container 25.5cm 4.34s 5.2 255g

Cylindrical Perforated 28cm 6s 6.5s 210gm

37 | P a g e
 basket

Bottle 20cm 8.5s 9.1s 157gm

Spherical Ball 21cm 8.4s 8.58s 70gm
Wallet Wallet 23cm 4.72s 5.4s 72gm

Calculator Calculator 22cm 4.85s 5.73s 97gm

Tester Tester 6cm 11.5s 12.12s 50gm

Objects like the wallet and calculator, which are flat and relatively lightweight, were grasped and
released faster (4.72 s and 5.4 s for the wallet; 4.85 s and 5.73 s for the calculator), reflecting their
favorable geometry for stable gripping. Interestingly, the smallest and lightest object—the tester—
took the longest time to manipulate (11.5 s to grab and 12.12 s to release), highlighting the
challenge of handling very small items with limited surface contact. These results indicate that the
robotic arm performs best with medium-sized, flat objects and faces challenges with round, small,
or slick-surfaced items.

Figure 4.1: Grab and release times for various objects

Figure 4.1 shows the grab and release times for various objects handled by the flexible robotic
arm. Objects like the Tester and Bottle required the longest times due to small size or unstable
shape, while flatter objects like the Wallet and Container were handled more quickly, indicating

38 | P a g e
better grip compatibility with regular geometries.

4.3 Degree of Freedom

The robotic arm operates in the X-Y plane and is constructed using 4 servo-driven rotational joints.
Each joint contributes one degree of freedom by rotating around the vertical (Z) axis. This gives
the arm 4 total degrees of freedom, enabling precise control of both the position and orientation of
the end-effector in 2D space.

Figure 4.2: Kinematic Diagram of a Multi-Link Robotic Arm

Table 4.2 outlines the joint configuration of the flexible robotic arm, focusing on the type of joint,
the motion it allows, and its corresponding Degree of Freedom (DOF). The robotic arm consists
of four joints, all of which are revolute joints—a common type in robotic mechanisms.

39 | P a g e
 Table 4.2: Degree of freedom of the prototype

Joint Type Motion DOF

1 Revolute Rotation around Z- 1

axis
2 Revolute Rotation around Z- 1
axis
3 Revolute Rotation around Z- 1
axis
4 Revolute Rotation around Z- 1
axis

Each revolute joint facilitates rotational motion around the Z-axis, contributing one degree of
freedom per joint. Specifically, Joint 1, Joint 2, Joint 3, and Joint 4 each enable independent
rotation about the Z-axis, providing a total of 4 degrees of freedom. This setup allows the arm to
perform versatile planar and spatial maneuvers while maintaining mechanical simplicity and
precise control.

4.4 Discussion

The developed robotic arm consists of four revolute joints, each actuated by servo motors, allowing
precise rotational movement primarily around the vertical (Z) axis. This configuration enables the
arm to perform movements within a two-dimensional planar workspace, suitable for a variety of
pick-and-place tasks. The mechanical structure of the arm is designed to be lightweight, utilizing
materials such as PLA (Polylactic Acid) for rigid components, ensuring not only ease of movement
but also cost-effectiveness in fabrication. Through a series of experimental trials, the robotic arm
was tested for its ability to grasp and release objects of different shapes and sizes, including
rectangular boxes, cylindrical containers, and spherical balls. The performance results indicated
that the arm handled mid-sized, regularly shaped objects such as wallets and plastic containers
most efficiently, showing stable grip and accurate placement. However, the system faced some

40 | P a g e
challenges when manipulating smaller or less stable objects, such as electronic testers, due to
limitations in grip precision and control.

The maximum payload capacity of the arm was found to be approximately 375 grams, a constraint
largely influenced by the torque output of the servo motor at the base joint, which bears the highest
mechanical load. This payload limit defines the practical range of applications for the arm,
particularly in scenarios involving light to moderate weight objects. On the control side, the robotic
arm employed a wireless communication system using HC-05 Bluetooth modules configured in a
master-slave arrangement, which facilitated real-time remote operation with minimal latency. The
communication link proved to be reliable, maintaining consistent actuation signals and smooth
motion execution throughout the tests. Overall, the combination of a modular, lightweight
mechanical design with an effective wireless control system demonstrates the arm’s capability for
real-world applications that prioritize affordability, flexibility, and ease of operation without the
need for complex wiring or tethered control. These attributes make it particularly suitable for
educational demonstrations, light industrial handling, and assistive robotics where adaptability and
wireless control are important.

5
6
7
8
9
10
11
12
13
14
15
16
17
18

41 | P a g e
 Chapter 5
Conclusions and Future Recommendations

5.1 Conclusion

This project successfully demonstrates that a flexible, low-cost robotic arm can be developed using
modular 3D-printed joints and Bluetooth-based wireless control. The system met its functional
goals through reliable object manipulation and remote operation, validating its suitability for real-
world, small-scale applications.

 The robotic arm was successfully built using affordable components and 3D-printed joints,
confirming the practicality of modular and low-cost design.
 The HC-05 Bluetooth-based system allowed reliable remote control with low latency,
increasing operational flexibility in constrained or remote environments.
 The arm, with four degrees of freedom, showed consistent performance in grasping and releasing
a variety of object types, particularly regular-shaped mid-weight items.
 The project met its intended objectives in terms of mechanical design, control, and basic
functionality.

5.2 Future Recommendations

While the current design and implementation of the flexible robotic arm meet the project’s primary
objectives, there remains significant potential for improvement to enhance functionality,
reliability, and autonomy. The current system is best suited for controlled environments and basic
manual operations. To make the robotic arm more capable of performing complex tasks in dynamic
or real-world settings, several advancements are recommended for future work.

 Incorporate AI and computer vision to enable object detection, tracking, and autonomous
decision-making for hands-free operation.
 Add more degrees of freedom or integrate soft material links to allow smoother, more
human-like motion and adaptability to irregular shapes.
 Use sensors (e.g., force, position, or tactile) to provide real-time feedback and improve the

42 | P a g e
 precision and safety of object handling.
 Evaluate performance in varied environmental conditions such as extreme temperatures or
radiation to ensure robustness in diverse settings.
 Upgrade to energy-efficient electronics and high-capacity batteries to extend operating
time and enable longer-term, field-based deployments.

43 | P a g e
 References

[1] G. C. Devol, Programmed article transfer, U.S. Patent 2,988,237, filed May 13, 1954, and
issued June 13, 1961.
[2] J. J. Craig, Introduction to Robotics: Mechanics and Control, 3rd ed. Upper Saddle River,
NJ, USA: Pearson, 2005.
[3] M. W. Spong, S. Hutchinson, and M. Vidyasagar, Robot Modeling and Control. Hoboken,
NJ, USA: Wiley, 2006.
[4] B. Siciliano and O. Khatib, Eds., Springer Handbook of Robotics, 2nd ed. Cham,
Switzerland: Springer, 2016.
[5] Intuitive Surgical, da Vinci Surgical System, White Paper, 2000.
[6] J. Burgner-Kahrs, D. C. Rucker, and H. Choset, “Continuum robots for medical
applications,” IEEE Trans. Robot., vol. 31, no. 6, pp. 1261–1280, Dec. 2015.
[7] L. L. Howell, S. P. Magleby, and B. M. Olsen, Handbook of Compliant Mechanisms.
Hoboken, NJ, USA: Wiley, 2013.
[8] X. Chen, Y. Liu, and M. Tan, “Wireless control strategies for soft continuum robots,” IEEE
Trans. Robot., vol. 37, no. 4, pp. 1125–1138, Aug. 2021.
[9] M. R. Sarker, J. Park, and M. S. Alam, “5G URLLC for industrial automation: Challenges
and case studies,” IEEE Trans. Ind. Informat., vol. 18, no. 4, pp. 2546–2557, Apr. 2022.
[10] Y. Zhang and J. Chen, “Additive manufacturing of multifunctional robotic components,”
Nature Mater., vol. 21, no. 4, pp. 412–418, Apr. 2022.
[11] Airbus, “Aerospace-grade validation of robotic systems for orbital servicing,” White Paper,
2023.
[12] D. E. Goldberg, “Genetic algorithms in reinforcement learning and robotics,” Robot. Auton.
Syst., vol. 15, no. 1–2, pp. 13–24, July 1995.
[13] MIT-Harvard Research Team, “Distributed intelligence in flexible robotic arms,” Research
Brief, 2024.
[14] ETH Zurich, “Variable stiffness actuation in continuum manipulators,” Technical Report,
2023.
[15] H. Luo, J. Xu, S. Li, H. Liang, Y. Chen, C. Xia and X. Wang, "D3-ARM: High-
Dynamic, Dexterous, and Fully Decoupled Cable Driven Robotic Arm," IEEE.

44 | P a g e
[16] J. Burgner-Kahrs, D. C. Rucker, and H. Choset,"A Continuum Robotics: Theoretical
Foundations, Applications, and Challenges,"IEEE Transactions on Robotics, vol. 31,
no. 6, pp. 1154–1171, Dec. 2015.
[17] Q. Zhou, Y. Lin, and Z. Wang, "A Variable-Stiffness Cable-Driven Robotic Arm for
Compliant Interaction," IEEE/ASME Trans. Mechatronics, vol. 26, no. 4, pp. 2021–
2032, Aug. 2021.
[18] Berkelman and Z. Ma, "A Compact 2-DOF Cable-Driven Robotic Arm for Confined
Spaces," IEEE Transactions on Robotics, vol. 32, no. 3, pp. 650–659, June 2016.
[19] Y. Tang, X. Liu, J. Zhao, L. Wang, and Q. Zhang, "Design and Control of a
Lightweight Cable-Driven Arm for Space Applications," Aerospace Science and
[20] X. Yang, J. Li, and H. Wang, "High-Precision Modeling and Control of Flexible Arms
Using Cable Transmission," Mechanism and Machine Theory, vol. 165, p. 104440,
2022.
[21] R. M. Murray, "A Cable-Driven Parallel Mechanism for a Flexible Surgical
Manipulator," IEEE Trans. Robot., vol. 32, no. 4, pp. 890–902, Aug. 2016.
[22] K. Xu, Y. Li, and H. Liu, "Design and Control of a Cable-Driven Continuum Robot
for Minimally Invasive Surgery," IEEE Trans. Mechatronics, vol. 23, no. 2, pp. 580–
589, Apr. 2018.
[23] J. Chen, M. Zhao, and L. Zhou, "Design of a Modular Cable-Driven Flexible
Manipulator," IEEE Access, vol. 8, pp. 112345–112356, 2020.
[24] H. Kim and S. Park, "Cable Tension Control for Improved Precision in Flexible Robot
Arms," IEEE Trans. Ind. Electron., vol. 65, no. 7, pp. 5574–5583, July 2018.
[25] A. Torres and R. Murphy, "Dexterous Cable-Driven Robotic Manipulators: Design
and Control," IEEE Robot. Autom. Mag., vol. 26, no. 3, pp. 64–76, Sept. 2019.
[26] F. Li, X. Xu, and G. Hu, "Design and Analysis of a Flexible Manipulator for Remote
Operations," IEEE Trans. Ind. Electron., vol. 66, no. 9, pp. 7142–7151, Sept. 2019.
[27] M. Savastano, F. Ruggiero, and B. Siciliano, "Modeling and Control of a Cable-
Driven Rigid-Link Manipulator," IEEE Trans. Robot., vol. 36, no. 4, pp. 1042–1053,
Aug. 2020.
[28] C. Wang, S. Hu, and Z. Cao, "Development of a Rigid Cable-Driven Manipulator for
Assembly Tasks," IEEE Trans. Autom. Sci. Eng., vol. 18, no. 1, pp. 322–332, Jan.

45 | P a g e
 2021.
[29] Y. Sun, W. Xu, and J. Wang, "Feedback Control System for a Flexible Robotic
Arm," IEEE Trans. Ind. Informat., vol. 15, no. 8, pp. 4485–4494, Aug. 2019.
[30] X. Liu and Y. Chen, "Adaptive Control of a Flexible Cable-Driven Robot Arm,"
IEEE Trans. Control Syst. Technol., vol. 28, no. 5, pp. 2043–2052, Sept. 2020.
[31] K. Xu, Y. Lin, and M. Yu, "Cable-Driven Lightweight Arm for Medical
Teleoperation," IEEE Access, vol. 9, pp. 55478–55488, 2021.
[32] J. Zhao and T. Li, "Hybrid Stiffness Control in Cable-Driven Robot Arms,"
IEEE Trans. Ind. Electron., vol. 69, no. 2, pp. 1678–1687, Feb. 2022.
[33] Y. Chen, D. Guo, and F. Zhang, "Compact Cable-Driven Robotic Arm Design for
Precision Tasks," IEEE Trans. Mechatronics, vol. 27, no. 1, pp. 138–149, Feb. 2022.
[34] D. S. Park, H. Lee, and J. Kim, "Integrated Sensor Feedback in Cable-Driven Arms
for Enhanced Flexibility," IEEE Trans. Ind. Informat., vol. 18, no. 6, pp. 3921–3931,
June 2022.
[35] L. Rucker and H. Choset, "Real-Time Control of Continuum Robots Using Model
Predictive Control," IEEE Trans. Robot., vol. 34, no. 5, pp. 1234–1244, Oct. 2018.
[36] L. Cassidy, “Arduino Uno Basics: Beginner’s Guide to Getting Started,” Flux.ai.
https://www.flux.ai/p/blog/arduino-uno-basics-beginners-guide-to-getting-started.
[37] “Arduino Uno Rev3,” Arduino Official Store. https://store.arduino.cc/products/arduino-
uno-rev3.
[38] “Arduino Uno R3 - Solarbotics Ltd.,” Solarbotics Ltd., Apr. 11, 2024.
https://www.solarbotics.com/product/50450.
[39] “Arduino UNO R3 CH340G ATmega328(SMD),” MakerBazar.in.
https://makerbazar.in/products/arduino-uno-r3.
[40] Robocraze, “What is Ultrasonic Sensor: Working Principle & Applications,” Robocraze,
Sep. 10, 2022. https://robocraze.com/blogs/post/what-is-ultrasonic-sensor.
[41] Barai, Sayantan & Maiti, Saswata & Modak, Arnab & Roy, Pallobi & Poyen Phd, Faruk.
(2018). Smart Weather Station 2018 Chapter I INTRODUCTION.
[42] Robu.in, “What is Arduino Pro Mini?,” Robu.in, [Online]. Available: https://robu.in/what-
is-arduino-pro-mini/. [Accessed: Jun. 20, 2025.
[43] "HC-05 Bluetooth Module," Components101, [Online]. Available:

46 | P a g e
 https://components101.com/wireless/hc-05-bluetooth-module. [Accessed: Jun. 20, 2025].
[44] 5V Single Channel Relay Module," Components101, [Online]. Available:
https://components101.com/switches/5v-single-channel-relay-module-pinout-features-
applications-working-datasheet. [Accessed: Jun. 20, 2025].
[45] What is a Perfboard? - Definition, Types, and Uses," Circuit Digest, [Online]. Available:
https://circuitdigest.com/tutorial/what-is-a-perfboard. [Accessed: Jun. 20, 2025].
[46] "Tower Pro MG90S Mini Digital Servo Motor," Robu.in, [Online]. Available:
https://robu.in/product/mg90s-metal-gear-servo-motor/. [Accessed: Jun. 20, 2025].

47 | P a g e
 Appendix
A: Micro-controller Program files

#include<SoftwareSerial.h>

SoftwareSerial btSerial(12, 13); // RX | TX

struct PacketData
{
byte Value1;
byte Value2;
};

int value;

PacketData data;

#define S1 3
#define S2 4

void setup()
{
pinMode(S1, INPUT);
pinMode(S2, INPUT);

btSerial.begin(9600);
Serial.begin(9600);
}

void loop()
{

48 | P a g e
 data.Value1 = digitalRead(S1);
data.Value2 = digitalRead(S2);

String dataString;
dataString = dataString
+ data.Value1 + ","
+ data.Value2 + "\n";

btSerial.print(dataString);
Serial.println(dataString);
delay(50);
}
B: Project Management and Budgeting
Resource Allocation:
B1. Hardware Components

Item Quantity Purpose

Arduino Uno R3 1 Main microcontroller for

control logic
Arduino Pro Mini 1 Secondary controller for
wireless module
HC-05 Bluetooth Module 2 Wireless communication
(master-slave setup)
MG996R Servo Motors 4 Actuation of 4-DOF robotic
arm joints
PLA/TPU Filament As needed 3D printing of arm segments
and gripper
Li-ion Batteries (3.7V) 5 Power supply for motors and
electronics
Buck Converters 3 Voltage regulation (5V/6V
for components)
Perfboard 1 Circuit assembly and stability

49 | P a g e
B2. Software/Tools

Resource Purpose

Arduino IDE Programming microcontrollers

Fusion 360 3D modeling of robotic arm parts

Proteus Circuit design and simulation

B3. Human Resources

 Team Members: 2 (Primary roles: Mechanical design, electronics, programming)
 Supervisor: Guidance on technical and research aspects.

B4. Cost analysis

Refer Table 3.2 in section 3.7(page 36) for the cost analysis of the project.

Justification:
 Cost-Efficiency: Prioritized low-cost, modular components (e.g., 3D-printed PLA parts,
off-the-shelf servos).
 Wireless Control: Bluetooth modules (HC-05) chosen for affordability and ease of
integration.
 Power Management: Buck converters ensure stable voltage for servos and electronics.

50 | P a g e
51 | P a g e
 C: Ethics, Teamwork, and Communication

C1 Plagiarism report and declaration

Figure C1: Similarity report.

52 | P a g e
 Research Ethics Compliance Checklist for Students
Decleration of Research Ethics Statement
☐ We have obtained the necessary permission from the relevant authorities to conduct this
research.
☐ No harm was inflicted on animals or any living beings during the conduct of this research. This
study strictly complies with all applicable laws regarding animal welfare and the protection of
personal data.
☐ We confirm that no data in this study has been manipulated, fabricated or falsified.
☐ We obtained necessary permission from relevant sources, clearly cite and acknowledge all
external sources.
☐ Others (if any)

Decleration of Conflict-of-interest Statement

Write the declaration of conflict-of-interest statement.

53 | P a g e
1. Student Name:
Roll No.:
Date:
Signature:

2. Student Name:
Roll No.:
Date:
Signature:

3. Supervisor Name:
Designation:
Date:
Signature:

54 | P a g e
Credit author statement

Md. Shahriar Mohtasim (Roll No.: 1802164): Conceptualization, Methodology,

Investigation, Formal analysis, Writing - Original Draft, Visualization.
Utpol Kumar Paul (Roll No.: 1802176): Conceptualization, Methodology, Investigation,
Formal analysis, Writing - Original Draft, Visualization.
Md. Golam Kibria (Assistant Professor, Department of Mechanical Engineering,
RUET): Conceptualization, Writing – review & editing, Supervision

55 | P a g e