Skip to content

About

Real-time computer vision project that controls a 5-finger robotic hand by mapping human gestures to servo angles using MediaPipe and Arduino

Resources

Stars

0 stars

Watchers

1 watching

Forks

Latest commit

 

History

7 Commits

Folders and files

Repository files navigation

Hand Tracking Robotic Hand Control

This project enables real-time control of a robotic hand using computer vision. It captures video from a webcam, detects hand landmarks using MediaPipe, calculates finger bending angles, and sends commands via serial to an Arduino to actuate 5 servo motors.

🛠️ Hardware Requirements

  • Arduino Board (Uno)
  • 5x Servo Motors (one for each finger)
  • Webcam
  • Power Supply (External 5V supply for servos)
  • Jumper wires & Breadboard

🔌 Wiring Configuration

Finger Arduino Pin
Thumb Pin 13
Index Pin 12
Middle Pin 11
Ring Pin 10
Pinky Pin 9

Circuit Diagram

Here is a wiring example showing two servos connected to an external power supply and an Arduino Uno. This setup should be replicated for all five servos. Rotating Spoon Arduino Schematic

💻 Software Installation

1. Python Environment

Install the required dependencies:

pip install opencv-python mediapipe pyserial tensorflow

About

Real-time computer vision project that controls a 5-finger robotic hand by mapping human gestures to servo angles using MediaPipe and Arduino

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages