A Raspberry Pi-based smart door opener that uses computer vision to recognize hand gesture sequences for secure door access. The system uses MediaPipe for real-time finger counting and controls a stepper motor via serial communication to a Seeed XIAO ESP32-C6.

• Hand Gesture Recognition: Uses MediaPipe to detect and count fingers in real-time
• Sequence-based Security: Requires a specific finger count sequence to unlock (default: 0, 1, 0, 5)
• Visual Feedback: NeoPixel LED ring provides status indication and feedback
• Stepper Motor Control: Serial communication with ESP32-C6 for precise door mechanism control
• Session Statistics: Tracks usage and provides summary reports

• Raspberry Pi 4 (recommended) or Pi 3B+
• USB webcam
• MicroSD card (32GB+ recommended)
• Power supply (5V 3A)

• WS2812B LED ring (20 LEDs recommended)
• Connected to GPIO 18 (configurable)

Wiring:

LED Ring -> Raspberry Pi
VCC      -> 5V (Pin 2)
GND      -> GND (Pin 6)
DIN      -> GPIO 18 (Pin 12)

• Seeed XIAO ESP32-C6 development board
• DRV8825 stepper motor driver
• Stepper motor (NEMA 17 recommended)
• 12V power supply for motor driver

ESP32-C6 Wiring:

ESP32-C6 -> DRV8825 Driver
D10      -> STEP
D9       -> DIR
D0       -> ENABLE
3.3V     -> VDD (logic power)
GND      -> GND

DRV8825 -> Motor/Power
VMOT     -> 12V+ (motor power)
GND      -> 12V- & logic GND
A1/A2    -> Motor coil A
B1/B2    -> Motor coil B

Serial Connection:

Raspberry Pi -> ESP32-C6
GND         -> GND
GPIO 14 (TX) -> RX (D7)
GPIO 15 (RX) -> TX (D6)

Alternative USB Connection:

• Connect ESP32-C6 via USB-C cable
• Default port: /dev/ttyACM0

The 3D/ folder contains STL files for 3D printable components:

• Board Holder.3mf: Mounting case for the ESP32 and DRV8825
• Camera Cover - Plate.3mf & Camera Cover - Top.3mf: Protective housing for the camera
• Door Gear.3mf: Door mechanism gear component
• Holder - Motor Holder.3mf: Mounting bracket for the stepper motor
• Motor Gear.3mf: Motor-side gear for the door mechanism

Print these components before assembly to ensure proper mounting and protection of your hardware.

# Update Raspberry Pi OS
sudo apt update && sudo apt upgrade -y

# Install system dependencies
sudo apt install git

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

# Restart shell or source profile
source ~/.bashrc

# Clone repository
git clone <repository-url>
cd door-pi

# Install Python dependencies with uv
uv sync

# Install PlatformIO Core
curl -fsSL -o get-platformio.py https://raw.githubusercontent.com/platformio/platformio-core-installer/master/get-platformio.py
python3 get-platformio.py

# Add to PATH (add to ~/.bashrc for permanent)
export PATH=$PATH:~/.platformio/penv/bin

# Navigate to firmware directory
cd Firmware

# Build firmware
pio run

# Upload firmware (ESP32-C6 connected via USB)
pio run --target upload

# Optional: Monitor serial output
pio device monitor --port /dev/ttyACM0 --baud 9600

# Add user to gpio group
sudo usermod -a -G gpio $USER

# Enable serial communication (if using GPIO serial)
sudo raspi-config
# Navigate to: Interface Options -> Serial Port
# Disable login shell over serial: No
# Enable serial port hardware: Yes

Edit main.py to change the unlock sequence:

self.DOOR_SEQUENCE = [0, 1, 0, 5]  # Change to your desired sequence

Modify the port in DoorLib/motor_control.py:

def turn_on(port='/dev/ttyACM0', baudrate=9600):  # Change port as needed

Adjust LED configuration in DoorLib/led_control.py:

def __init__(self, pixel_pin=board.D18, num_pixels=20, brightness=1.0):

Adjust stepper motor parameters in Firmware/src/main.cpp:

#define STEP 50              // Steps to move (adjustable)
#define WAITING_TIME 5000    // Hold time in milliseconds

# Activate virtual environment
uv run python main.py

1. Idle State: Blue chasing light animation on LED ring
2. Hand Detection: LEDs show current finger count with colored regions
3. Sequence Entry: Enter the finger count sequence (default: fist → 1 finger → fist → 5 fingers)
4. Success: Green flash, door unlocks, stepper motor rotates and holds
5. Failure: LEDs return to idle, sequence resets

The ESP32-C6 accepts these serial commands:

• #on: Move stepper motor forward, hold for 5 seconds, then disable
• #off: Rewind stepper motor to original position and disable

• Blue chase: Idle/waiting for input
• Colored regions: Current finger count display
• Cyan segments: Sequence progress indication
• Green flash: Successful unlock
• Return to idle: Wrong sequence (no error indication)

# Add new dependencies
uv add package-name

# Run specific commands
uv run python main.py
uv run pytest  # if tests exist

# Build without upload
pio run -e seeed_xiao_esp32c6

# Clean build
pio run --target clean

# Update libraries
pio lib update

• Change the default sequence before deployment
• Consider adding time delays between sequence attempts
• Monitor the system logs for unauthorized access attempts
• Ensure proper physical security of the Raspberry Pi and ESP32-C6
• Use appropriate power ratings for your stepper motor

# List USB cameras
lsusb | grep -i camera

# Test camera with OpenCV
python3 -c "import cv2; cap = cv2.VideoCapture(0); print('Camera available:', cap.isOpened()); cap.release()"

# List available serial ports
ls /dev/tty*

# Test serial connection to ESP32-C6
minicom -D /dev/ttyACM0 -b 9600

# Check GPIO permissions
sudo chown root:gpio /dev/gpiomem
sudo chmod g+rw /dev/gpiomem

# Check if device is detected
lsusb | grep -i esp

# Reset ESP32-C6 permissions
sudo chmod 666 /dev/ttyACM0

# Reinstall dependencies
uv sync --reinstall

This project is open source. Please use responsibly and ensure proper security measures for your specific use case.

Contributions are welcome! Please ensure all changes maintain the security and reliability of the system.