Mark 1 Hardware Stack

Frame & Powertrain

[WIP]

Vision Systems

Raspberry Pi 5 (8GB)

Arducam OV9281 USB

MicroSD Card (64GB)

USB-C Data Cable

---

RainGuard

Hardware BoM [TBD]

Compute & Control

• AI Computer: NVIDIA Jetson Orin Nano (8GB)
• Flight Controller: Pixhawk 6C (or Cube Orange) running ArduPilot
• GPS: Holybro Micro M10 GPS (or equivalent GNSS)
• Telemetry: Internal MAVLink stream to Jetson via UART

Vision & Connectivity

• Primary Camera: Intel RealSense D435i (Depth + RGB) OR High-Quality CSI Camera (IMX477)
• Wi-Fi Module: Intel AX210 M.2 or Alfa AWUS036ACM (for Smartphone Hotspot)
• RC Transmitter: RadioMaster Pocket (ELRS) - Required for Emergency Kill Switch

Propulsion & Frame

• Frame: Custom X-Class Carbon Fiber (1000mm+ wheelbase to clear umbrella)
• Motors: Low KV / High Torque (400KV - 600KV)
• Propellers: 15-18 inch Carbon Fiber
• ESC: 4-in-1 60A+ ESC (BLHeli_32)
• Power: 6S LiPo Battery (10,000mAh+) with AS150 Anti-Spark Connectors

Umbrella Mechanism

• Canopy: Lightweight wind-vented umbrella (stripped of handle)
• Mount: 2-Axis Gimbal (Gravity stabilized) to keep umbrella upright during flight pitch

Wiring Diagram

Wiring diagrams and pinout maps are located in the hardware/schematics directory.

Connection Overview

• Jetson <-> Pixhawk: UART (TELEM2) for MAVLink communication
• Camera <-> Jetson: USB 3.0 or CSI Ribbon Cable
• Power: PDB to ESCs and Voltage Regulator (5V/12V) for Jetson

Installation

Hardware Setup

1. Frame Assembly: Assemble the drone frame using files in hardware/cad. Ensure arms are long enough so propellers do not overlap with the umbrella diameter. ALEX!: Get an "X-Class" or "Cinelifter" style geometry, not a standard DJI shape!
2. Mounting: Secure the Jetson and Pixhawk to the center plate with vibration damping.
3. Umbrella: Attach the gimbal mechanism to the top plate.
4. Power: Solder AS150 connectors to the PDB to prevent sparking.

Software Setup

Prerequisites

• Python 3.8 or higher
• ROS 2 (Humble recommended)
• MAVSDK or DroneKit
• Git LFS (Required for AI models)

Installing Dependencies

# Clone the repository
git clone https://github.com/The1AndOnlyAlex/RainGuard.git
cd RainGuard

# Initialize Git LFS (Crucial for .pt models)
git lfs install
git lfs pull

# Install Python dependencies
pip install -r requirements.txt

Web App Setup (Smartphone Control)

This project uses a Flask-based web interface hosted on the Jetson to control the drone.

1. Network: The Jetson creates a Hotspot named Aegis-Link.
2. Connect: Connect your smartphone to Aegis-Link.
3. Launch: Open browser to http://192.168.50.1:5000 (default IP).

Usage

1. Power On

Connect battery. Wait for ESC initiation beeps. Wait 60s for Jetson to boot and start the Aegis-Link hotspot.

2. Emergency Override

ALWAYS have the RadioMaster Pocket in hand.

• Switch SA: Arm/Disarm
• Switch SD: EMERGENCY KILL (Motor Interlock)

3. Running the System (Dev Mode)

# Launch the full stack (Vision + Control + Web Server)
python src/main_system.py

4. Running with ROS2

# Launch the autonomous drone system
ros2 launch rainguard aegis.launch.py

Project Structure

RainGuard/
├── hardware/
│   ├── cad/             # STL/STEP files for mounts & frame
│   └── schematics/      # Wiring diagrams
├── src/
│   ├── vision/          # YOLOv8 & DeepSORT tracking logic
│   ├── control/         # MAVLink flight control scripts
│   └── web_interface/   # Flask app for smartphone control
├── models/              # AI Model weights (*.pt files)
├── docs/                # Setup guides & calibration
└── requirements.txt     # Python dependencies

License

See LICENSE file for details.