Drone Simulation Controller

This project provides a Python-based controller for operating a drone in AirSim simulation environment, typically integrated with Unreal Engine 4 (UE4). The script allows real-time keyboard control of a virtual drone for testing and development purposes.

Features

• Real-time keyboard control of drone movement
• Intuitive control scheme:
  • W / S: Move forward/backward
  • A / D: Move left/right
  • ↑ / ↓ (Arrow keys): Ascend/descend
  • Q / E: Rotate left/right
  • ESC: Land and exit
  • C: Capture and save current camera view
  • Y: Send current camera view to YOLO object detection model
• Live camera feed display during flight
• Automatic image saving with sequential numbering
• YOLO object detection integration (requires separate YOLO server)
• Smooth drone movement with configurable speed settings
• Safe takeoff and landing procedures

Prerequisites

Before running this project, ensure you have:

1. Unreal Engine 4 installed
2. AirSim plugin installed and configured with UE4
3. Python 3.10 installed (virtual environment recommended)
4. Required Python packages:
   • airsim
   • keyboard
   • opencv-python
   • numpy
   • requests

Installation

1. Clone this repository:

   git clone <repository-url>
   cd drone

2. Create a virtual environment:

   python -m venv venv

3. Activate the virtual environment:

   • On Windows:

     venv\Scripts\activate

   • On macOS/Linux:

     source venv/bin/activate

4. Install required packages:

   pip install airsim keyboard opencv-python numpy requests

Usage

1. Start your AirSim simulation environment (UE4 project with AirSim plugin)

2. (Optional) Start your YOLO object detection server on a separate port (local or cloud)

3. Update the YOLO_SERVER_URL in drone.py to point to your YOLO server if different from default

4. Run the drone controller:

   python drone.py

5. The drone will automatically takeoff and await your commands

6. Use the keyboard controls listed above to maneuver the drone

7. Press C to capture and save the current camera view at any time

8. Press Y to send the current camera view to the YOLO model for object detection

9. Press Q to close the live camera feed window

10. Press ESC to land the drone and exit the program

Images will be saved in the captured_images directory with sequential numbering.

Configuration

You can adjust the drone's behavior by modifying these parameters in drone.py:

• DRONE_SPEED: Movement speed in meters per second (default: 5)
• ROTATION_SPEED: Rotation speed in degrees per second (default: 45)
• IMAGE_DIR: Directory where captured images are saved (default: "captured_images")
• YOLO_SERVER_URL: URL of the YOLO object detection server (default: "http://localhost:5000/yolo")

Project Structure

.
├── drone.py          # Main drone control script
├── README.md         # This file
├── .gitignore        # Git ignore rules
├── venv/             # Python virtual environment (ignored by Git)
└── captured_images/  # Directory for saved drone camera images (created automatically)

Safety Notes

• The drone automatically lands when the program exits
• The virtual environment is excluded from Git tracking
• Ensure your AirSim simulation is running before executing the script

Contributing

Feel free to fork this project and submit pull requests with improvements or bug fixes.