Hand Gesture Controlled Gripper

This project uses MediaPipe to detect hand gestures from a webcam feed and controls a Dynamixel servo motor accordingly. When the system detects an open or closed hand, it sends a command to the motor to open or close a gripper in real-time.

---

Features

• Real-time hand gesture recognition using MediaPipe
• Integration with Dynamixel SDK for servo motor control
• Current-based position control for safe and compliant motion
• Smooth visual feedback via OpenCV

---

Demo Gestures

Gesture	Gripper Action
✋ Open Hand	Gripper Opens
✊ Closed Fist	Gripper Closes

---

Dependencies

Install dependencies via pip:

pip install opencv-python mediapipe numpy

Additionally, install the Dynamixel SDK (Python):

cd ~/DynamixelSDK-3.7.31/python
python3 setup.py install

---

How to Run

1. Connect your Dynamixel servo to your system via USB (default port: /dev/ttyUSB0).
2. Ensure the Dynamixel ID and port are correct in new_algorithm.py.
3. Run the script:

python new_algorithm.py

4. Show your hand to the webcam. Open or close it to control the gripper.

---

Configuration

You can tune the following parameters in new_algorithm.py:

• DXL_ID – Motor ID (default: 1)
• DEVICENAME – Serial port (default: /dev/ttyUSB0)
• GOAL_CURRENT_VALUE – Current limit for compliance
• DXL_MINIMUM_POSITION_VALUE, DXL_MAXIMUM_POSITION_VALUE – Gripper open/close position range
• threshold – Distance threshold for gesture classification

---

File Structure

hand_gesture_gripper_control/
├── new_algorithm.py       # Main control script
├── README.md              # Project documentation
└── requirements.txt       # Optional Python dependencies list

Acknowledgments

• MediaPipe
• ROBOTIS Dynamixel SDK