Below is a video showcase of the entire project on YouTube!

PalmPilot is a computer vision-powered, hand gesture-controlled Arduino car. Using a webcam and AI, you can drive, steer, and stop the car with simple hand gestures—no physical remote required!

• Control car movement (forward, backward, left, right, stop) using hand gestures
• Real-time gesture recognition with computer vision (MediaPipe + OpenCV)
• Arduino-based motor control via serial communication
• Plug-and-play: just connect your Arduino and webcam

1. The Python script uses OpenCV and MediaPipe to detect hand gestures from your webcam.
2. Recognized gestures are mapped to car commands (e.g., open palm = forward, fist = stop).
3. Commands are sent to the Arduino over serial.
4. The Arduino receives commands and drives the motors accordingly.

• Arduino board (Uno, Nano, etc.)
• Motor driver (L298N or similar)
• Webcam
• Motors and wheels
• USB cable

• Python 3.x
• OpenCV (cv2)
• MediaPipe
• PySerial
• Arduino IDE

1. Clone the repository:

   git clone https://github.com/Rian-Sen/PalmPilot.git

2. Install Python dependencies:

   pip install opencv-python mediapipe pyserial

3. Connect your Arduino and upload arduino.ino using the Arduino IDE.
4. Update the serial port in serial_connection.py to match your system.
5. Run the Python script:

   python serial_connection.py

• Open palm: Move forward
• Fist: Stop
• Middle finger up: Move backward
• Index finger up: Turn right
• Index + middle finger up: Turn left

• serial_connection.py — Computer vision and gesture recognition logic
• arduino.ino — Arduino code for motor control
• LICENSE — MIT License
• README.md — Project documentation
• Computer Vision Car Report.pdf — Technical report

MIT License © 2025 Rian Sen Majumder