Design and Implementation of an Arduino-Based Mobile Robotic Arm with Omnidirectional Mobility for High-Precision Industrial Automation

Efficiency and precision have made automation a part of modern industrial plants. In this era the robotic arm is a symbol of technology, can do intricate tasks with precision. Automation of industrial processes using robotic arms has increased productivity by reducing manual labor, increasing speed and minimizing human error [1]. This project designs and implements a sophisticated robotic arm control system for industrial automation and precision manufacturing. It addresses the need for more efficiency and accuracy in today’s industrial setting where traditional manufacturing methods—manual labor or basic equipment—are being challenged by their limitations: human error, slow processing and high operational cost [2]. Industries are fast tracking the adoption of automation technology and robotic arms are playing a key role in replicating human dexterity across industrial tasks [3].

However, integrating robotic arms in precision manufacturing is not without its challenges. The complexity of controlling their movements often leads to inconsistencies and errors which defeats the purpose of precision they are supposed to deliver [4]. Conventional control systems lacking accuracy and adaptability struggle to meet the dynamic demands of modern industrial environment. This work aims to bridge this gap by integrating advanced control techniques, artificial intelligence (AI) to enhance precision, flexibility and productivity of robotic arms. Building on existing research that explores various control methods—fuzzy logic, PID control and neural networks—this work will optimize robotic arm operation [5]. Moreover, real-time feedback systems and sensor fusion will improve accuracy and reliability. The goal is to develop a robotic arm system that can do precise pick-and-place movement and affordable and easy to operate so it can be a viable and efficient solution for various industrial applications. The problem this work is trying to address is the optimal control for precision manufacturing and industrial automation using robotic arms. Current control methods are not providing the required accuracy and adaptability to the industry demands. The limitations of traditional control methods—slow real-time response, execution errors, high production cost and material waste—highlight the need for a more refined approach [6]. While neural control, fuzzy logic and PID control have been explored along with advancements in sensor technology and feedback systems, existing technologies may still not meet the precise requirements of industrial automation. Therefore, this work will develop an intuitive software interface for robotic arm control along with prototyping tools that will simplify setup, programming and operation so it will be affordable and efficient.

In pursuit of these goals this work has three main objectives: first to develop a robotic arm system that can do precise pick-and-place; second to design an affordable and user-friendly robotic arm; and third to create a control system that can integrate the robotic arm into existing industrial processes. The scope of this work is to fabricate a 5-axis robotic arm controlled by a smartphone. To achieve this, we need to have a comprehensive understanding and application of various technical domains including C programming, Arduino Mega microcontroller, Bluetooth communication and MIT App Inventor mobile application development. Furthermore, the testing of the prototype will enhance technical knowledge and provide valuable insights into the robotic arm functionality and reliability in real world industrial environment so that it can be suitable for various industrial automation applications.

This work is also driven by the growing need for automation solutions that can bridge the gap between human dexterity and machine precision. In many industries from electronics to pharmaceuticals the need for high quality output is critical. Traditional robotic systems struggle with fine movements for intricate tasks and hence bottlenecks and quality control issues. This work aims to address these challenges by developing a control system that not only enhances robotic arm precision but also adapts to various manufacturing scenarios. The integration of advanced algorithms including machine learning will enable the robotic arm to learn from its environment and adjust its movements in real time reducing errors and increasing efficiency. This adaptability is crucial for industries where product designs and manufacturing processes change frequently and require rapid reconfiguration and minimal downtime.

Also, the work highlights the importance of accessibility and affordability in automation technology. While high end robotic systems have advanced features their cost and complexity limit their adoption by small and medium enterprises (SMEs). By developing a user friendly and cost-effective robotic arm control system this work aims to democratize automation and make it accessible to a wider range of industries. The use of off the shelf components like Arduino Mega microcontroller and open source software tools like MIT App Inventor will reduce the overall cost of the system. Also, the development of a smartphone interface will simplify the programming and operation of the robotic arm and eliminate the need of technical expertise. This will not only lower the barrier to entry for automation but also enable businesses to use advanced technology without significant capital investment.

But the bigger picture goes beyond the immediate benefits of efficiency and cost savings. By showing that advanced control techniques can be applied to affordable robotic systems, this work contributes to the advancement of industrial automation as a whole. The knowledge gained from this will inform the development of future robotic systems and lead to more sophisticated and adaptable automation solutions. Plus, the focus on user friendly interfaces and accessible technology aligns with the trend towards human-robot collaboration where robots work alongside humans to increase productivity and safety. By making automation more approachable and intuitive, this work fosters a more collaborative and efficient manufacturing environment, ultimately driving innovation and competitiveness across diverse industrial sectors.

The main contributions of the proposed research work are as follows:

• Incorporates mecanum wheels for full 360° mobility, enabling the robotic arm to navigate tight industrial settings—a significant improvement over traditional, fixed-position arms.

• Achieves up to 95% accuracy by calibrating MG996R and SG90 servos and employing Denavit–Hartenberg kinematic modeling, surpassing the 90% benchmark of many existing low-cost systems.

• Utilizes 3D-printed components and readily available Arduino-based hardware to create a cost-effective solution that is easy to reproduce and well-suited for small-to-medium manufacturing enterprises.

• Offers an MIT App Inventor–built interface with Bluetooth communication, simplifying robotic arm and base control while reducing the need for specialized technical expertise.

• Ensures stability through a segregated power supply and a 5-DOF structure, allowing easy upgrades—such as AI-driven vision—without requiring extensive redesign.

• Employs quantitative experiments to measure joint-position accuracy, motion speed, and power usage, verifying system robustness and identifying potential areas for further enhancement.

The proposed method is for a portable and precise robotic arm for industrial automation. It uses mecanum wheels for omnidirectional mobility to adapt to dynamic environments. The robotic arm is powered by MG996R and SG90 servo motors and controlled by an Arduino Mega microcontroller programmed in Embedded C. A Bluetooth module (HC-05) is used for wireless control via a mobile app developed using MIT App Inventor for users.

3.1 Mechanical design

The mechanical design of the robotic arm is focused on flexibility, precision and portability for industrial automation tasks. The robotic arm has 5 degrees of freedom (DOF) structure which can perform movements like rotating, extending and gripping so it is suitable for pick-and-place operations.

The robotic arm is built using 3D printed parts as shown in Figure 1 which is lightweight and cost effective without compromising the structural integrity. The modular design makes it easy to assemble and maintain and can be customized or replaced easily. Each joint is calibrated to have smooth and precise motion and servos provide the required torque for movement.

1.png

Figure 1. 3D printed parts

The mecanum wheels in the base of the robotic arm are a big innovation. These wheels allow the arm to move forward, backward, sideways and diagonally. This feature allows the system to move in tight and confined industrial spaces and more efficient. The wheels are powered by DC motors controlled by a motor driver shield for smooth and responsive movement.

3.2 Electrical system

The robotic arm’s electrical system is the backbone that ties everything together and gives you control. The design prioritizes efficiency, reliability and compatibility with the mechanical and software subsystems. The Arduino Mega 2560 microcontroller is the heart of the system which manages all the electrical components and executes the control algorithms. This microcontroller is chosen for its ample input/output pins, high processing capability and compatibility with various peripherals. It receives commands from the Bluetooth module and translates it into precise control signals for the motors.

Servo Motors

The robotic arm uses a combination of servo motors:

• MG996R servo motors for the waist, shoulder and elbow joints which requires higher torque to handle heavier loads and larger movements.

• SG90 servo motors for the wrist and gripper which provides fine control for delicate operations like gripping small objects.

These servos are powered by a separate power supply to avoid overloading the microcontroller.

The HC-05 Bluetooth module is used for wireless communication between the robotic arm and the user’s smartphone. This module allows real time command input so the user can control the arm remotely through a custom mobile app. It works up to 10 meters range and is reliable for typical industrial setup.

The whole system is powered by a lithium-ion battery pack, chosen for its high energy density and long life. The battery supplies power to the servo motors, DC motors and the Arduino Mega, so everything runs smoothly. A buck converter is used to regulate the voltage for sensitive components to prevent overvoltage or undervoltage. Figure 2 shows the hardware components.

2.png

Figure 2. Circuit diagram

3.3 Power consumption and runtime

During extended operation, the MG996R servo motors exhibited a noticeable increase in temperature. To evaluate thermal behavior, infrared temperature measurements were recorded at 5-minute intervals while the robotic arm executed continuous motion cycles for 30 minutes under moderate load conditions. The servo casing temperature gradually rose to approximately 55℃, particularly in the shoulder and elbow joints which are subjected to higher torque demands.

Although this temperature remains within the operational limit of MG996R servos (typically up to 60℃), sustained exposure near this threshold may lead to long-term performance degradation or motor failure. To mitigate this, two preventive strategies were applied:

1. Passive aluminum heat sinks were attached to the servo casing to improve heat dissipation.
2. The PWM signal was software-limited to a maximum of 70% duty cycle, which helped reduce thermal output without sacrificing motion smoothness.

These improvements significantly reduced the peak operating temperature by an estimated 6–8°C in repeated tests. For future versions, the integration of temperature sensors and active thermal control systems (e.g., mini fans or thermal shutdown logic) is recommended for enhanced reliability.

3.4 Programming the microcontroller

The Arduino Mega 2560 microcontroller is programmed in Embedded C within the Arduino IDE. The program controls all the hardware, so everything communicates and executes commands smoothly. Key features of the microcontroller software are:

• Initialization: The program initializes all the hardware components, servo motors, motor driver shield and Bluetooth module. It sets the default positions of the robotic arm joints and establishes connection with the mobile app.

• Command Processing: The microcontroller processes the commands received from the Bluetooth module. The commands are interpreted as specific actions like moving a joint, rotating the base or operating the gripper.

• Control Algorithms: The software implements control algorithms to ensure precise movements. For example, PWM (Pulse Width Modulation) signals are used to control the angle and speed of the servo motors and DC motors.

3.5 Mobile application

The mobile app developed using MIT App Inventor provides a user-friendly interface to control the robotic arm wirelessly. Key features of the app are:

• Graphical User Interface (GUI): The app has buttons and sliders to control the arm’s movements like adjusting joint angles, rotating the base and operating the gripper.

• Bluetooth Connectivity: The app connects to the HC-05 Bluetooth module to transmit commands in real time to the microcontroller.

• Preset Functions: The app has pre-programmed commands for common tasks like picking up an object or moving the arm to a specific position, to make user life easier.

In MIT App Inventor, the Blocks Editor allows users to design the logic of their mobile applications by connecting visual code blocks. Key types include Event Blocks, which trigger actions in response to user interactions or system events (e.g., sending a command when a button is clicked), and Control Blocks, which manage program flow through conditions and loops (e.g., using if...then to check Bluetooth connectivity before sending commands). Action Blocks perform specific tasks, like sending text via Bluetooth, while Variable Blocks store and manipulate data, such as joint angles or status updates. Together, these blocks create a user-friendly yet powerful framework for controlling the robotic arm and processing feedback efficiently as shown in Figure 3.

3a_.png

3b.png

Figure 3. Block editor

The Bluetooth-based control interface for the robotic arm was developed using MIT App Inventor, offering a simple yet effective user interface for controlling arm movement servo actuation.

Figure 4 shows  repository contains all the necessary blocks, UI layout, and deployment instructions, enabling other developers or researchers to replicate or modify the app as needed. The system architecture is modular and supports scalability, allowing integration of multiple robotic arms through unique device IDs and expanded control panels. This provides a pathway for future development of multi-robot coordination, cloud-based control features, and integration with Industry 4.0 frameworks such as IoT dashboards or remote diagnostics platforms.

4.png

Figure 4. Source code repository

3.6 Working flow

The Development of Robotic Arm Control System for Industrial Automation using Arduino involves several key stages, from component selection and design to control and integration of the system. The following Figure 5 describes the system’s working flow, explaining how the components interact to achieve the system's objectives.

5.png

Figure 5. Block diagram

The working flow is as follows:

1) Power Supply

The system is powered by an external power supply that provides the necessary voltage and current to the Arduino Mega, Bluetooth module, servo motors, and DC motors for the mecanum wheels.

2) User Input

User controls the robotic arm using a mobile app. The app sends commands via Bluetooth to the Arduino which interprets the commands and controls the motors accordingly.

3) Bluetooth Communication

HC-05 Bluetooth module establishes the wireless connection between the mobile app and the Arduino Mega. App sends commands to control the movement of the robotic arm and mecanum wheels.

4) Control System

Arduino Mega processes the received commands and sends the appropriate commands to the servo motors for the robotic arm and DC motors for the mecanum wheels. Arduino uses a predefined program written in Embedded C to control the servos movement and arm functions.

5) Servo Motors for Arm Movement

MG996R and SG90 servos are used to move the joints of the robotic arm. Servos are controlled by the Arduino based on the user input received through the mobile app.

6) Mecanum Wheels for Mobility

Motor control shield for the Arduino drives four DC motors which are connected to the mecanum wheels. These wheels allow the arm to move in all directions and provide portability to the system.

7) Task Execution

Once the system is controlled, the robotic arm can perform pick-and-place tasks with high precision.

3.7 Working kinematic modelling of the 5-DOF robotic arm using the Denavit-Hartenberg method

Kinematic model of the 5-degree-of-freedom (DOF) robotic arm is defined using the Denavit-Hartenberg (D-H) convention which is a systematic method for representing the spatial relationships between consecutive links of a robotic manipulator. The D-H method uses 4 parameters for each joint: joint angle θi, link offset di, link length ai, and twist angle αi. These parameters describe how one coordinate frame is related to the next in terms of translation and rotation. The transformation between each consecutive joint is calculated using a homogeneous transformation matrix based on these parameters.

The standard D-H parameter table for the 5-DOF robotic arm is in Table 2. The values of θi are the joint variables that change during operation, while the parameters di, ai, and αi are constants because they are defined by the physical structure of the robotic arm.

Table 2. Denavit-Hartenberg parametersfor the 5-DOF robotic arm

Table 2 is the structured framework for the forward kinematics of the robotic arm. The transformation matrices are calculated from these parameters so you can calculate the position and orientation of the end effector relative to the base.

3.8 Transformation matrix for each joint

Each joint in the robotic arm is represented by a homogeneous transformation matrix. The general transformation matrix between two consecutive joints, using the standard D-H convention, is expressed as:

$\mathrm{T}_{\mathrm{i}}^{\mathrm{i}-1}=\left[\begin{array}{cccc}\cos \theta_i & -\sin \theta_i \cos \alpha_i & \sin \theta_i \sin \alpha_i & \alpha_i \cos \theta_i \\ \sin \theta_i & \cos \theta_i \cos \alpha_i & -\cos \theta_i \sin \alpha_i & \alpha_i \sin \theta_i \\ 0 & \sin \alpha_i & \cos \alpha_i & d_i \\ 0 & 0 & 0 & 1\end{array}\right]$

This matrix defines the relative position and orientation of one joint to the previous one by combining rotational and translational transformations. By calculating the transformation matrices for all joints and multiplying them together you get the overall transformation from the base frame to the end effector.

The total transformation matrix for the robotic arm is:

$\mathrm{T}^5{ }_0=\mathrm{T}^1{ }_0 \mathrm{~T}^2{ }_1 \mathrm{~T}^3{ }_2 \mathrm{~T}^4{ }_3 \mathrm{~T}^5{ }_4$

where, each individual transformation matrix $\mathrm{T}_i^{\mathrm{i}-1}$ is derived from the respective D-H parameters of each joint. This final transformation matrix gives the complete kinematic description of the robotic arm and the exact location and orientation of the end effector in 3D space.

3.9 Inverse kinematics and trajectory analysis

To enable the robotic arm to reach specific spatial targets with precision, an inverse kinematics (IK) routine was implemented. The IK solution calculates the required joint angles based on the desired end-effector coordinates. For this project, a simplified geometric approach was applied using the Jacobian Transpose method, suitable for low-degree-of-freedom manipulators.

A 2-link planar approximation was used to derive joint angles for the shoulder and elbow axes. The equations below describe the trigonometric relationship between the end-effector position and link angles:

$\theta_2=\cos ^{-1}\left(\frac{x^2+y^2-L_1^2-L_2^2}{2 L_1 L_2}\right)$

$\theta_1=\tan ^{-1}-\left(\frac{y}{x}\right)-\tan ^1\left(\frac{L_2 \sin \left(\theta_2\right)}{L_1+L_2 \cos \left(\theta_2\right)}\right)$

where:

• x,y = Desired end-effector position in 2D plane
• L₁, L₂ = Lengths of the shoulder and elbow links
• θ₁ = Shoulder joint angle
• θ₂ = Elbow joint angle

This method was integrated into the control logic via Arduino and used to test various end-effector targets.

Table 3. Inverse kinematics test

Table 3 shows the angles were computed using real-time trigonometric functions, then mapped to the corresponding PWM signal to control servo rotation. Deviations were mainly due to servo backlash, mechanical tolerances, and floating-point limitations in microcontroller computation.

3.10 Bluetooth latency testing and its impact

To evaluate the responsiveness of the system’s wireless control, a latency test was conducted using the HC-05 Bluetooth module and summarized in the Table 4 below. A total of 50 command transmissions were issued from the mobile app to the Arduino-based control system, consisting of repeated ON/OFF toggles for the servo activation signal. Timestamp logging was used on both the sender (mobile) and receiver (microcontroller) ends to measure the delay.

Table 4. Bluetooth latency test

The average latency across all tests was 270 ms, with a peak delay recorded at 310 ms. This level of delay is within acceptable limits for low- and medium-speed industrial applications where precise timing is not mission-critical. However, in high-speed or time-sensitive operations, such latency could introduce cumulative delays and potential misalignment.

The developed Arduino-based mobile robotic arm with omnidirectional mobility successfully demonstrated the ability to perform pick-and-place operations with a high degree of flexibility and control. The integration of MG996R and SG90 servos enabled a 5-degree-of-freedom motion profile, while mecanum wheels ensured maneuverability across multiple directions. The Bluetooth-controlled mobile app offered a low-cost and user-friendly method for remote control.

The robotic system achieved an overall positioning accuracy of approximately 93% during physical testing, validating its potential for medium-precision industrial automation tasks. The system also exhibited good repeatability and stability during dynamic operations and various surface mobility tests.

The authors extend their appreciation to Universiti Teknikal Malaysia Melaka (UTeM) and to the Ministry of Higher Education of Malaysia (MOHE) for their support in this research.

The authors’ contributions are as follows: “Conceptualization, J.A.J.A and S.S.; methodology, J.A.J.A. and M.F.Y.; software, A.H.A.; validation, M.F.B.T.W. and S.S.; formal analysis, M.F.Y.; investigation, J.A.J.A; resources, A.H.A.; writing—original draft preparation, J.A.J.A and S.G.H.; writing—review and editing, S.G.H. and M.F.B.T.W.; funding acquisition, M.F.Y. and S.G.H. All authors have read and agreed to the published version of the manuscript.