Skip to content

Latest commit

 

History

History
183 lines (136 loc) · 6.59 KB

README.md

File metadata and controls

183 lines (136 loc) · 6.59 KB
Raw

Thermal Imaging Drone for Search and Rescue Operations

Cover Photo SAR Thermal Imaging Drone

Overview

This project introduces a thermal imaging drone designed for efficient search and rescue (SAR) operations. The system integrates a thermal camera with object detection and GPS tracking, enabling real-time location and detection of individuals in disaster zones.

Key Features

  • Thermal Imaging: FLIR Vue Pro R thermal camera detects heat signatures in real-time, even in low-visibility conditions.
  • Object Detection: Utilizes YOLOv5 for accurate and real-time identification of individuals.
  • Live GPS Tracking: Logs precise GPS coordinates of detected persons for easy localization.
  • Data Logging: Automatically saves detected locations as images and CSV files.
  • User-Friendly: Compatible with QGroundControl and Mission Planner for intuitive operation.

Hardware Requirements

Component Specifications
Drone Frame TAROT X6 Hexacopter frame
Flight Controller Pixhawk Cube Orange Plus
Thermal Camera FLIR Vue Pro R
Gimbal Custom 3D-printed two-axis gimbal
Airunit & Controller Herelink 1.1
Battery Tattu 22.2V 25C 6S 22000mAh Lipo
GPS Module Here+ RTK GPS
Telemetry RFD 868 ux-IND

Software Requirements

  • Mission Planner: Ground control station software for configuration and operation.
  • QGroundControl: Real-time monitoring and data visualization.
  • YOLOv5: Object detection algorithm for human identification.
  • MySQL: Database to store GPS coordinates of detections.

Installation and Setup

1. Hardware Configuration

  • Refer the wiring diagram for connection
  • Connect the flight controller, GPS module, and telemetry components.
  • Test connections using Mission Planner and QGroundControl.
  • Mount and calibrate the thermal camera and gimbal.

For detailed assembly and configuration instructions, refer to the Hardware Setup and Configuration document.

2. Software Setup and Installation

Step 1: Clone the Repository

Clone the repository and navigate to the project folder:

git clone https://github.com/your-repository/Thermal-Imaging-Drone-SAR.git
cd Thermal-Imaging-Drone-SAR

Step 2: Software Installation

  • Install Mission Planner (Guide)

  • Install QGroundControl (Guide)

  • Set up YOLOv5 (Guide)

  • Install MySQL for GPS database creation. (Guide)

    Note: As part of the YOLOv5 setup, ensure all Python dependencies are installed using the requirements.txt file included in this repository. Run the following command after cloning the repository:

    pip install -r Software/requirements.txt

Step 3: Organize Code Files

Move the necessary scripts into their respective folders after installation:

  • YOLOv5 Files: 
    cp Software/detect.py ~/YOLOv5
cp Software/gps_function.py ~/YOLOv5
cp Software/plottime.py YOLOv5/
  • MAVLink Script: 
    cp Software/mavlink_gps_to_sql.py MAVLink/

Usage

Standard Operating Procedure (SOP)

Refer to the detailed SOP.md file for a comprehensive guide. Below is a quick summary:

  1. Pre-Flight Preparations:

    • Verify hardware connections and power on all systems.
    • Use Mission Planner to check GPS signal and system readiness.

  2. Run YOLOv5 Detection:

    • Start MAVLink GPS logging: 
      python3 MAVLink/mavlink_gps_to_sql.py
    • Start object detection: 
      python3 YOLOv5/detect.py --source rtsp://192.168.43.1:8554/fpv_stream --class 0

  3. Data Retrieval:

    • Access detected locations, images, and CSV files in the YOLOv5/detect/exp/ folder.

  4. Post-Flight Operations:

    • Safely land the drone and back up logs, images, and CSV files.

Testing and Results

Testing Summary

  • Indoor Testing: Verified YOLOv5 with webcam and streamed thermal feeds.
  • Outdoor Testing: Successfully detected humans at altitudes of 20m and 50m with live GPS tracking.

Results

  • Successfully detected and logged human locations in low-visibility conditions.
  • Thermal video streaming tested at altitudes of 20m and 40m with accurate detections.
  • Stored results as images and CSV files for further analysis.

Repository Structure

Thermal-Imaging-Drone-SAR/
├── Documentation/             # Guides for installation, hardware setup, SOP, and more
│   ├── SOP.md                 # Standard Operating Procedure
│   ├── Hardware_Setup_and_Configuration.md
│   └── Installation_Guides/
├── Hardware/                  # CAD models, components list, and wiring diagrams
│   ├── Components_List.csv
│   ├── 3D_Models/
│   │   └── Gimbal.stl
│   └── Wiring_Diagram.jpg
├── Software/                  # Python scripts and dependencies
│   ├── detect.py
│   ├── gps_function.py
│   ├── plottime.py
│   ├── mavlink_gps_to_sql.py
│   ├── requirements.txt       # Python dependencies
├── YOLOv5/                    # YOLOv5-related files and weights
├── MAVLink/                   # MAVLink-related scripts and configurations
│   └── mavlink_gps_to_sql.py
├── Results/                   # Output results, including images and CSV files
│   ├── Detected_Locations.csv
│   ├── Detected_Images/
└── README.md                  # Main project documentation

Contributors

  • Srinivasan Ravindran - Project Head
  • Shafeek PM - Progarm Coodinator
  • Sidharth Mohan Nair - Project Lead
  • Anooja SK & Jobin J - Software Integration
  • Sandeep S & Anwar Sherief - Hardware Design

License

This project is licensed under the MIT License.

References