A 1.85kg autonomous quadcopter platform built for GPS waypoint navigation, position hold, return-to-launch, and Raspberry Pi MAVLink control.
This repo contains everything needed to build and program an autonomous quadcopter from scratch -- hardware specs, wiring guides, ArduCopter configuration, and Python flight control software.
- GPS waypoint navigation via DroneKit-Python
- MAVLink control from Raspberry Pi 4 companion computer
- GPS position hold and altitude hold
- Automatic return-to-launch on low battery or signal loss
- 12-15 minute hover flight time on 3000mAh 4S LiPo
- 1.89:1 thrust-to-weight ratio
- Computer vision ready (future expansion)
In Development
Completed:
- ArduCopter V4.6.3 firmware flashed
- ESC and motor configuration
- GPS module connected and tested
- Frame design finalized
- Carbon-fiber arm/motor assemblies built and dry-fit
- ESC power wiring bench-tested
In Progress:
- Sensor calibrations
- Frame assembly
- Raspberry Pi MAVLink integration
- Companion computer and GPS deck mounting
Upcoming:
- First test flight
- Autonomous mission testing
- Computer vision integration
The current hardware build is being documented as it moves from frame dry-fit to powered bench testing. All tests shown here were performed with propellers removed.
Arm/motor assemblies: carbon-fiber arms with 3D-printed motor mounts, mounted brushless motors, and motor phase leads staged at the inner ends.
Frame dry-fit: four arm assemblies installed into the center plate before the electronics stack and cable management are finalized.
Center hub wiring: motor leads routed toward the center deck before final ESC connection and strain relief.
Companion computer stack: Raspberry Pi, Matek F405-Wing V2, BN-880 GPS, and jumper wiring staged for MAVLink/GPS bench testing.
ESC power wiring bench test: XT60 input leads, motor bullet connectors, and LiPo checker used while validating solder joints and polarity.
Powered electronics stack test video: 12-second MP4 showing the powered center stack, flight controller LEDs, GPS, Raspberry Pi, XT60 connection, and LiPo checker during a no-prop bench test.
| Spec | Value |
|---|---|
| Total Weight | 1.85 kg (all-up) |
| Thrust-to-Weight | 1.89:1 |
| Flight Time (Hover) | 12-15 minutes |
| Flight Time (Aggressive) | 8-10 minutes |
| Max Thrust | 3500g (875g per motor) |
| Frame Size | 450mm diagonal |
| Propeller Size | 11x7 inches |
| GPS Accuracy | +/- 2-3 meters |
| Wind Resistance | Stable in 5-10 mph |
| Component | Model |
|---|---|
| Flight Controller | Matek F405-Wing V2 (STM32 F405, 168MHz) |
| ESC | Diatone Mamba F40 4-in-1 (40A, Dshot600) |
| Motors | D2830-12 Brushless x4 (850 KV) |
| Battery | 3000mAh 4S LiPo (14.8V, 60C) |
| GPS | Beitian BN-880 (Ublox M8N + compass) |
| Companion Computer | Raspberry Pi 4 (2-4GB) |
| Frame | Custom carbon fiber rod (450mm, 10x 18mm rods, 3D-printed motor mounts) |
| Firmware | ArduCopter V4.6.3 |
Total Cost: ~$250-$450 USD
Autonomous-Drone/
├── README.md
├── LICENSE
├── config/
│ └── arducopter-params.param # Flight controller parameters
├── docs/
│ ├── assets/
│ │ └── build-progress/ # Build photos and bench-test video
│ ├── build-guide.md # Step-by-step assembly
│ ├── hardware-specs.md # Detailed component specs
│ ├── flight-controller-setup.md # ArduCopter configuration
│ ├── mavlink-setup.md # Raspberry Pi integration
│ ├── calibration-guide.md # Sensor calibration procedures
│ ├── troubleshooting.md # Common issues and fixes
│ └── safety.md # Safety protocols
├── hardware/
│ ├── bill-of-materials.md # Full parts list with costs
│ ├── frame/
│ │ └── frame-assembly.md # Frame build instructions
│ └── wiring-diagrams/
│ └── wiring-guide.md # Complete wiring reference
└── software/
├── requirements.txt
├── setup.sh # Raspberry Pi setup script
├── drone_controller.py # Core DroneKit controller class
├── waypoint_mission.py # Autonomous waypoint missions
└── monitoring_dashboard.py # Real-time telemetry display
See the Build Guide for full step-by-step instructions. The Bill of Materials has the complete parts list.
Connect the Matek F405-Wing V2 via USB and flash ArduCopter V4.6.3 using QGroundControl or Mission Planner. Then load the parameter file:
config/arducopter-params.param
See Flight Controller Setup for details.
Follow the Calibration Guide:
- Accelerometer (6-position)
- Compass (outdoor, all orientations)
- ESC throttle range
# Clone this repo on the Pi
git clone https://github.com/rishith-c/Autonomous-Drone.git
cd Autonomous-Drone
# Run the setup script
chmod +x software/setup.sh
./software/setup.sh
# Or install manually
pip3 install -r software/requirements.txtSee MAVLink Setup for UART wiring and configuration.
from dronekit import connect
vehicle = connect('/dev/ttyAMA0', wait_ready=True, baud=921600)
print(f"GPS: {vehicle.gps_0}")
print(f"Battery: {vehicle.battery}")
print(f"Mode: {vehicle.mode.name}")
vehicle.close()from software.drone_controller import DroneController
drone = DroneController()
drone.arm()
drone.takeoff(5)
drone.goto(37.12345, -121.12345, 10)
drone.rtl()
drone.close()[3000mAh 4S LiPo 14.8V]
|
[XT60 Connector]
|
[Mamba F40 ESC]
/ | | \
M1 M2 M3 M4
|
[VBAT/GND to FC]
|
[Matek F405-Wing V2]
|
[5V BEC to peripherals]
├── GPS Module (5V)
├── Raspberry Pi (separate 5V BEC or USB power bank)
└── Other peripherals
[Matek FC - ArduCopter]
├── ESC (Dshot600) --> Motors
├── GPS Module (UART3, 115200 baud)
├── Compass (I2C, integrated with GPS)
├── Raspberry Pi (UART2, MAVLink, 921600 baud)
└── RC Receiver (optional, UART/SBUS)
User / Mission Planner
|
Raspberry Pi (Python / DroneKit)
| MAVLink commands
Flight Controller (ArduCopter)
| Sensor fusion: GPS, IMU, Baro
ESC (motor speed commands)
|
4x Brushless Motors
FRONT
M3 CCW M1 CW
\ /
[FC]
/ \
M2 CW M4 CCW
M1: Front Right (CW)
M2: Rear Left (CW)
M3: Front Left (CCW)
M4: Rear Right (CCW)
| Document | Description |
|---|---|
| Build Guide | Complete assembly from start to finish |
| Hardware Specs | Detailed specs for every component |
| Bill of Materials | Full parts list with costs |
| Wiring Guide | All wiring connections |
| Frame Assembly | Frame construction |
| Flight Controller Setup | ArduCopter parameters |
| MAVLink Setup | Raspberry Pi integration |
| Calibration Guide | Sensor calibration |
| Troubleshooting | Common problems and solutions |
| Safety | Safety protocols and checklists |
Short Term (1-3 months)
- FPV camera for first-person view
- RC receiver for manual backup
- Obstacle avoidance sensors
- Telemetry radio for longer range
Medium Term (3-6 months)
- Computer vision with OpenCV
- Object tracking and following
- Precision landing using visual markers
- Advanced path planning
Long Term (6+ months)
- Machine learning integration
- Autonomous inspection missions
- Multi-drone swarm coordination
- Indoor flight with alternative positioning
- Payload delivery system
- ArduPilot Documentation
- DroneKit-Python Docs
- MAVLink Protocol
- Matek F405-Wing V2 Docs
- QGroundControl
This project is licensed under the MIT License. See LICENSE for details.