Slide 1: Title Slide

Script: "Good morning/afternoon, everyone. My name is Mai Gia Huy, and I am here to
present my senior project titled 'Robot Arms for Object Classification Utilizing QR Codes.'

Slide 2: Table of Contents

Script: "Today, we will cover the following sections:

1. Introduction
2. Project Management
3. Literature Review
4. Methodology
5. Expected Results
6. Conclusion and Future Work By the end of this presentation, you will have a clear understanding of our
project objectives, the steps we took to achieve them, and the potential impact of our work."

Slide 3: Introduction

Script: "Let's begin with the introduction.

1. Problem: Rapid industrialization and modernization are driving the need for advanced technology in
production facilities. Despite these advancements, the deployment of robots in Vietnam is limited due
to high costs, making robotic technology unaffordable for many businesses. This limits the ability of
local industries to enhance productivity and quality. Our project aims to address this issue by localizing
robotic technology, which can enhance productivity, improve quality, and reduce costs for businesses.
2. Background: Our project involves creating a mechanized device that mimics the functionality of a
human hand. This device, consisting of interconnected segments (links) with joints for movement, can
execute complex tasks flexibly, which is essential in automation and manufacturing. The system
captures images of a conveyor belt, identifies objects by subtracting the current frame from a static
backdrop, and analyzes features for classification using QR codes. The methods we use include:

 Object Detection: Calculates pixel coordinates for robot retrieval.

 Object Tracking: Continuously updates coordinates across frames to ensure accurate tracking of
objects.
 Object Classification: Uses shape-based recognition and QR codes to classify objects efficiently.

1. Goal & Objective:

 Goal: The goal of our project is to develop an autonomous robotic system for object classification using
robot arms and QR code technology. This system will automate the classification process to improve
efficiency and accuracy in various applications, such as manufacturing and logistics.
 Objectives:

o Objective 1: Hardware Development

 Task 1.1: Develop circuit boards that will control the various components of the
robot.
 Task 1.2: Program the circuits to ensure precise control and coordination of the
robot's movements.
o Objective 2: Software Development

 Task 2.1: Develop an image recognition system capable of accurately detecting and
classifying objects based on QR codes.
 Task 2.2: Create a customer database to store information about the classified
objects and their destinations.
  Task 2.3: Develop the system interface to provide users with an easy-to-use control
panel for operating the robotic system."

Slide 4: Project Management

Script: "Now, let's look at the design specifications and management of our project.

1. Components & Budget: The hardware components used in our project include:

 Arduino UNO R3: A microcontroller board used for controlling the robot arm. It acts as the brain of the
robot, processing inputs from sensors and executing control commands.
 Robot Arm MK2: The main robotic arm mechanism with multiple degrees of freedom, allowing it to
perform a wide range of movements and tasks.
 Stepper Motors: Used for precise control of the robot arm movements. They ensure the robot arm can
position itself accurately to pick up and place objects.
 CNC Shield V3 and A4988 Driver: Used to control the stepper motors. The CNC shield provides the
necessary interface between the Arduino and the motors, while the A4988 driver ensures smooth and
precise motor control.
 Power Supply: Provides the necessary power to the entire system, ensuring all components operate
reliably.
 Geared Motors and Limit Switches: Ensure accurate positioning and movement control, preventing the
robot arm from exceeding its operational limits.
 Relay Module and Vacuum Pressure Pump: Used for gripping and releasing objects. The vacuum pump
creates suction to pick up objects, while the relay module controls its operation.
 Webcam: Captures images of the objects for QR code recognition. It provides real-time visual data for
the image recognition system.
 Conveyor System: Moves objects into the robot arm's working area, ensuring a steady flow of items to
be classified. The total budget for these components is 4,034,000 VND.

1. Project Timeline: Our project is divided into several phases:

 Phase 1: Planning: Define project objectives, requirements, and timeline.

 Phase 2: System Design: Design the hardware and software architecture of the system.
 Phase 3: Hardware Assembly: Assemble the robot arm and other hardware components.
 Phase 4: Software Development: Develop the software for image recognition, control algorithms, and
the user interface.
 Phase 5: Hardware-Software Integration: Integrate the hardware and software components to create
a functional system.
 Phase 6: System Testing: Test the system to ensure it operates as expected and make any necessary
adjustments.
 Phase 7: Deployment and Final Adjustments: Deploy the system in a real-world environment and
make final adjustments to optimize performance.
 Phase 8: Project Review and Closure: Review the project outcomes, document the results, and close
the project."

Slide 5: Literature Review

Script: "Let's move on to the literature review, where we highlight key contributions to our
project.

1. Denavit & Hartenberg (1955):

 Contribution: Introduced the Denavit-Hartenberg (D-H) notation for robot arm kinematics, a standard
method for assigning parameters to each joint. These parameters include offset (d), joint angle (θ),
angle (α), and link length (a).
  Application: This notation is used in our project for precise gripper positioning, forward kinematics
solution, trajectory planning, and collision avoidance during robot arm movement.

1. Craig (1989):

 Contribution: Provided a comprehensive foundation for understanding robot manipulator control,

exploring various control techniques for accurate trajectory following.
 Application: Feedback control techniques from Craig's work ensure accurate gripper positioning and
continuous monitoring and adjustments for precise QR code scanning.

1. Mamdani & Assilian (1975):

 Contribution: Introduced fuzzy logic for handling uncertainties in real-world systems, which is effective
for addressing variations in lighting, partial occlusions, and sensor noise.
 Application: Fuzzy logic improves QR code detection and decoding accuracy, allows dynamic
adjustments in image processing parameters, and increases the robustness of our vision system.

1. Eraghi et al. (2014):

 Contribution: Compared various path planning algorithms (HCTNav, A*, and Dijkstra's) in grid maps,
highlighting their strengths and weaknesses for specific applications.
 Application: These path planning algorithms help in efficient object manipulation, minimizing the risk
of damage, and optimizing the scanning sequence for multiple QR codes."

Slide 6: Methodology

Script: "Next, let's delve into the methodology of our project.

1. System Overview: Our system captures real-time images using a webcam. Algorithms detect and
classify objects by QR codes. The system calculates object coordinates and controls the robotic arm via
Arduino. The Arduino uses A4988 drivers and a CNC Shield V3 for stepper motors. Conveyor belt speed
is controlled by PID and Fuzzy PID algorithms.
2. Hardware Development:

 Developing Circuit Boards: The webcam captures and processes images. The system uses object
detection, QR code recognition, and coordinate calculation. Inverse kinematics algorithms control the
robot arm. The Arduino drives stepper motors for precise positioning.
 Robot Arm Hardware Connection: Components include a 12V DC power supply, stepper motors
connected to CNC Shield V3, and limit switches for initial reference positions.
 Conveyor and Pump Connection: Conveyor motor control is independent via a relay module with a
speed control circuit. The pump motor operates in a single direction, controlled by a relay.

1. Programming Features:

 The robot initializes to the home position.

 Identifies and classifies objects using QR codes.
 Moves objects to designated positions. Unclassified objects remain unpicked.

1. Software Development:

 Video Frame Analysis: Continuously analyzes video frames from a camera to detect objects.
 Background Subtraction: Identifies objects by comparing new frames to a reference image of an empty
belt.
 Region of Interest (ROI): Focuses on a specific area to improve accuracy and efficiency.
 Noise Reduction: Removes unwanted variations for reliable background subtraction.
  Object Detection: Determines the object center and contour, using QR code information for
categorization."

Slide 7: Expected Results

Script: "As for the expected results, we aimed to achieve the following:

1. Hardware: An efficiently functioning robotic arm with accurate movement and control, capable of
picking and placing objects based on QR code data.
2. Software: A reliable object detection and classification system that accurately processes QR codes and
interacts seamlessly with the hardware.
3. Overall: Demonstrated enhancement in automation and productivity in manufacturing processes,
providing a practical solution for industries looking to adopt cost-effective robotic technology. We
anticipate that our system will significantly reduce the need for manual labor in repetitive tasks,
thereby increasing overall efficiency and productivity."

Slide 8: Conclusion and Future Work

Script: "In conclusion, we successfully developed a robotic arm system capable of classifying
objects using QR codes. This project contributes to the localization of robotic technology and
offers a cost-effective solution for manufacturing automation. While our system has
demonstrated promising results, there is always room for improvement. Future research can
focus on enhancing the system's accuracy, expanding its capabilities to handle more complex
tasks, and exploring new applications in other areas of the industry. By continuing to innovate
and improve, we can further advance the adoption of robotic technology in local industries.

1. Project Achievements:

 Significant milestones in robotic arm control systems.

 Deep understanding of robot systems and necessary equipment.
 Effective control of robot movements using Arduino and Python.
 Near-perfect object detection accuracy on the conveyor belt.
 95% object identification rate.
 Successful coordinate conversion between the image frame and the robot's coordinate system.
 A strong foundation for further development.

1. Problems and Limits:

 Sensitivity to light changes and overheating.

 Motor heat management.
 User interface software stability.

1. Future Work:

 Addition of color, shape, and size identification capabilities.

 Upgrade robot arm and conveyor belt motors for durability and smoother operation.
 Replace limit switches with modern optical sensors for improved joint calibration accuracy."

Slide 9: Questions

Script: "Thank you for your attention. I am now open to any questions you may have about
the project."