MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION GRADUATION THESIS COMPUTER ENGINEERING TECHNOLOGY DESIGN AND IMPLEMENTATION OF A FIREFIGHTING ROBOT INSTRUCTOR: DR.TRUONG NGOC SON STUDENT: LE QUOC THANH SKL012487 Ho Chi Minh City, JUNE, 2024 HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION FACULTY FOR HIGH QUALITY TRAINING GRADUATION PROJECT DESIGN AND IMPLEMENTATION OF A FIREFIGHTING ROBOT LÊ QUỐC THANH Student ID: 19119049 Major: COMPUTER ENGINEERING TECHNOLOGY Advisor: Assoc. TRƯƠNG NGỌC SƠN Ho Chi Minh City, June 2023 THE SOCIALIST REPUBLIC OF VIETNAM Independence – Freedom– Happiness -------- Ho Chi Minh City, June 26, 2023 GRADUATION PROJECT ASSIGNMENT Student name: Lê Quốc Thanh Student ID: 19119049 Major: Computer Engineering Technology Class: 19119CLA Advisor: Assoc. Trương Ngọc Sơn Phone number: 0931085929 Date of assignment: 30 September, 2023 Date of submission: 26 June, 2023 1. Project title: Design and implementation of a firefighting robot 2.
Initial materials provided by the advisor: Documents, such as papers and books, that pertain to AI play a role in controlling hardware devices. Content of the project: - Analyze and invetigate the challenges of the project. - Research technical specifications and operating principles based on the theoretical basis of hardware components. - Propose the model and apply and deploy the model.
System design, principle diagram, block diagram. - Select data, prepare data and label objects (clean data, create object detection data). - System configuration and design hardware. - Test run, check, evaluate and adjust.
- Conduct report writing. Final product: A miniature robot, affixed to the ceiling, is capable of identifying and detecting fires, and it can initiate extinguishing measures using sprinklers. CHAIR OF THE PROGRAM ADVISOR (Sign with full name) (Sign with full name) THE SOCIALIST REPUBLIC OF VIETNAM Independence – Freedom– Happiness -------- Ho Chi Minh City, June 26, 2023 ADVISOR’S EVALUATION SHEET Student name: Lê Quốc Thanh Student ID: 19119049 Major: Computer Engineering Technology Project title: Design and implementation of a firefighting robot Advisor: Assoc. Trương Ngọc Sơn EVALUATION 1.
Content of the project:. Approval for oral defense? (Approved or denied). ) Ho Chi Minh City, June …, 2023 ADVISOR (Sign with full name) THE SOCIALIST REPUBLIC OF VIETNAM Independence – Freedom– Happiness -------- Ho Chi Minh City, June 26, 2023 PRE-DEFENSE EVALUATION SHEET Student name: Lê Quốc Thanh Student ID: 19119049 Major: Computer Engineering Technology Project title: Design and implementation of a firefighting robot Name of Reviewer:. Content and workload of the project.
Approval for oral defense? (Approved or denied). ) Ho Chi Minh City, June …, 2023 REVIEWER (Sign with full name) THE SOCIALIST REPUBLIC OF VIETNAM Independence – Freedom– Happiness -------- Ho Chi Minh City, June 26, 2023 EVALUATION SHEET OF DEFENSE COMMITTEE MEMBER Student name: Lê Quốc Thanh Student ID: 19119049 Major: Computer Engineering Technology Project title: Design and implementation of a firefighting robot Name of Defense Committee Member::. Content and workload of the project. ) Ho Chi Minh City, June …, 2023 COMMITTEE MEMBER (Sign with full name) Disclaimer We hereby declare that this is the final report, "Design and implementation of a firefighting robot".
The simulations and study findings are accurate and were carried out entirely under the direction of the instructor, Assoc. TRUONG NGOC SON. The report does not duplicate any other sources either. Additionally, the paper includes a variety of cited and carefully labeled reference materials.
Before the department, faculty, and school, we would like to fully accept responsibility for this promise. Student Lê Quốc Thanh i Acknowledgements First, we would like to express my deepest gratitude to the School Board of the Ho Chi Minh City University of Technology and Education, as well as the Faculty for High Quality Training, for establishing wonderful conditions for me to pursue our project. In addition, we would also like to express our sincere thanks to the Head of the Department, Assoc. Truong Ngoc Son, who always closely follows the learning situation and encourages and creates development opportunities for each generation of students.
Last but not least, we cannot prevent mistakes due to a lack of knowledge and implementation time. We welcome your feedback and suggestions to help us improve this topic. Many thanks for all your help and regards. Student Lê Quốc Thanh ii Table of Contents Disclaimer.
ii List of Figures. v List of Abbreviations. viii Chapter 1: INTRODUCTION. Designing a Firefighting Robot with Raspberry Pi and Arduino Integration.
Research methods applied to robot and scope of robot research. Research methods applied to robot. Scope of robot research. 3 Chapter 2: LITERATURE REVIEW.
Predictive tracking of an object .2 Pan–tilt control system .3 Finding the Object Realtime Position. The Overview of Firefighting Robot. 7 Chapter 3: DESIGN AND MANUFACTURE OF FIREFIGHTING ROBOT. Main goals and technical objectives of firefighting robots.
Block diagram of firefighting robot. System design and layout of Firefighting robot .1 Fire detection system .1 Hardware system design for fire detection .2 Software system design for fire detection. Pan-tilt control for fire detection tracking .1 Hardware system design for pan-tilt control .2 Software system design for object tracking using Pan-Tilt mechanism. 34 iii Chapter 4: EXPERIMENT RESULTS AND DISCUSSION.
System Operation and results .1 Fire object detection fire detection .2 Pan and tilt controls .2 Review and evaluation. 44 Chapter 5: CONCLUSION AND FUTURE WORKS. 47 iv List of Figures Figure 2. YOLO network architecture diagram.
Method for pan-tilt camera calibration. 1 Robot block diagram. 2 Controller CNC Shield V4. 3 Raspberry pi 4 model B.
4 Raspberry pi 4 schematic. 5 Arducam Fisheye low light USB camera. Stepper motor structure. Transmitted in the form of packets of UART.
9 Block diagram of the robot's operating process. 10 Algorithm processing flow chart on raspberry pi. 11 Performance of yolov8 compared to other versions. 12 Upload the dataset to roboflow.
13 Yolov8 model is ready on roboflow. 14 Collaboration between roboflow and docker. 15 Command set up docker on raspberry pi. 16 Command pull docker.
17 Command run docker with CPU. 18 Command run docker with CPU. 19 Results when running docker. 20 Results obtained after training.
21 Calculate bounding box detection. 22 Sample Detection of burning objects. Check the FPS implementation. 24 Robot skeleton structure.
25 Complete frame structure diagram. Firefighting robot moves along a frame structure. 27 Arduino CNC Shield V4 GRBL circuit. 28 Arduino CNC Shield V4 GRBL schematic.
29 TMC2209 Stepper Driver module. 30 TMC2209 Stepper Driver module schematic. 32 Arduino nano V3 schematic. Algorithm processing flow chart on Arduino nano.
Calculate coordinates on image. Results of the robot's operation. Results Of Yolov8's training Process. 38 Results Of Moving Pan and Tilt.
44 vi List of Abbreviations AI Artificial intelligence CV Computer vision YOLO You Only Look Once UART Universal Asynchronous Receiver/Transmitter SDA Serial Data SCL Serial Clock MSB Most significant bit UART Universal Asynchronous Receiver/Transmitter LSB Least significant bit GPU Graphics processing unit CPU Central Processing Unit GPIO General Purpose Input/Output vii Abstract In response to growing concerns about sudden fires and the need for rapid intervention, the development of firefighting robots has become urgent. That's why we decided to carry out the topic "Design and implementation of a firefighting robot". Our project aims to innovate and apply AI automation within the robot , our firefighting robot project seeks to address the urgency of fire detection and mitigation. The concept originated from the realization that unforeseen fire outbreaks can pose significant threats, necessitating a proactive and automated response.Our approach involved the integration of sophisticated image processing methods with cutting-edge machine learning models.
So that robots designed to find a fire, before it rages out of control, can one day work with fire-fighters greatly reducing the risk of injury to victims. The challenge in this process lies in accurately directing the nozzle to the intended ignition point. We initiated the project with a fire detection model utilizing the YOLO model, known for its capacity to analyze and detect objects with a high degree of accuracy. Additionally, we implemented an automation processing model employing to enhance precision during navigation.The ultimate outcome is impressive, demonstrating exceptional smoothness and accuracy.
viii Chapter 1: INTRODUCTION In the first chapter, we will explain the reason for choosing the project and clarify the project's research object, methodology and objectives of the project, in addition to the process and how it works. Introduction It is our responsibility as Electrical Engineers to design and build an embedded system that can automatically detect and extinguish fires. Also aims to reduce air pollution. According to statistics from the Ministry of Public Security, from 2001 to 2022, Vietnam will have nearly 60,000 fires and explosions, killing 1,910 people, injuring more than 4,400 people, and causing great financial damage.
Many house fires start when someone is sleeping or not home. With the invention of such a device, people and property can be saved at greater expense with relatively little damage caused by fire. Firefighting using robot firefighting is the best today. Following in the footsteps of leading technologists such as HOPE Technick, Fire Robot Technology.
We will create a small automatic firefighting robot that can navigate at very fast speeds and track objects, find the burning object and then extinguish it with a direct high-pressure water jet.Firefighting robots are designed to search for interior fires small floor plan of a house of a specific size, extinguish the fire with water, then return to the original position. The fire detection prediction models put into use are relatively error-free, they will not overreact in non-fire simulations. Artificial intelligence (AI) is becoming increasingly popular and is affecting many aspects of daily life. Computer vision (CV) is a branch of artificial intelligence that includes the collection, processing, analysis, and recognition of digital images.
yolo is one of the most stable vision models available today, with flexibility and ease of access and deployment on embedded systems. Together with the precision of automatic control devices, the implementation of the project becomes more feasible. Therefore, we decided to choose the topic "Firefighting Robot". Designing a Firefighting Robot with Raspberry Pi and Arduino Integration The research objective of the thesis is the design and implementation of firefighting robot has the following functions: Use Raspberry Pi 4 to process images, calculate object positions and create signal coordinates transmitted via UART communication standard.
Use the Arducam Fisheye Low Light USB Camera for Computer, to recognize fire. Use an Arduino Nano to control the pan and tilt through module driver TMC2209. Use two module driver TMC2209 to control the pan angle and the tilt angle. Track and verify advancement using the computer.
1 Calculate and design the most optimal hardware models Considering assessing and refining the model. The expected result will be a robot capable of detecting sudden fires and promptly preventing them from extinguishing them. Objective It is crucial to analyze and assess the accuracy, speed of processing, and performance of models for fire recognition on embedded systems. Evaluate and analyze system functions to select appropriate hardware that brings high performance and efficiency.
Research methods applied to robot and scope of robot research 1. Research methods applied to robot In order to gain a deeper grasp of the issue's implementation, we performed research on the research topics, which made problem-solving easier. The topics we looked into were as follows: The Raspberry Pi 4 Model B has garnered significant popularity among developers. Engineered as a compact yet potent embedded computer, it serves a myriad of purposes, making it ideal for learning and testing.
Its versatility and high compatibility enable seamless access and deployment of both AI applications and control of other hardware.
