Skip to content

ROH-Gen2-Demos Suitable for ROH-AP001/AP002

Demo Acquisition

  • Method 1: Click this link to download

  • Method 2: Visit ohand_serial_sdk GitHub page to download the latest version

  • Method 3: Obtain via clone: 

    git clone https://github.com/oymotion/roh_gen2_demos

Demo Operation Tutorial (Windows System)

ROHand's Demo is developed based on Python scripts. Below is the usage tutorial for Windows systems:

  • Currently supported ROHand models include ROH-AP001 and ROH-AP002 series.

  • The Demo currently offers three control methods: vision control, glove control, and gForce myoelectric armband control.

1. Install python3.12.4

Please ensure the host username is in English before installation

Click here get python3.12.4 from official website
or Click here to download the installer

Double-click the exe to open the Python installer, and be sure to check the option shown in the figure below before installation

Figure1

Click Install Now

After installation is complete, you can open the command line to check if the installation was successful 

python --version

If the installation is successful, the console will display Python 3.12.4

2. Install CH341SER Driver

Ckick here to download CH341SER driver

Double-click the exe to open the CH341SER installer

Click Install to proceed

3. Create virtual environment (optional)

Using a virtual environment ensures that installing demo dependencies does not affect other environments 

pip install virtualenv

# Select any location where you want to configure the virtual environment
cd path/to/your/folder

# Create a virtual environment, "name" is the virtual environment name and can be customized
virtualenv name

# Activate virtual environment
.\name\Scripts\activate.bat

4. Install Required Dependencies

4.1 glove_ctrled_rohand

The glove-controlled dexterous hand demo requires bleak-winrt library, so additional installation of vs_BuildTools is necessary

Click here to download vs_BuildTools

Double-click th exe file to open theVisual Studio Installer

Check the required installation packages as shown in the figure (mandatory)，others can be selected according to your needs

Figure2

After checking the options, click Install to proceed

After installation, enter in the console: 

cd /path/to/demo/glove_ctrled_rohand
pip install -r requirements.txt

After installation is complete, enter 

python glove_ctrled_hand.py

Can start the demo

The program will automatically detect whether you are using a wired or wireless glove.

Important Notes:

  • You need to select single-dimensional force or three-dimensional force based on the dexterous hand model you purchased. If you purchased AP001, choose TACS_DOT_MATRIX; otherwise, select TACS_3D_FORCE, then choose left or right hand.

  • The glove requires calibration before use. After running the program, repeatedly perform hand opening-closing-opening movements until calibration is complete. During calibration, try to stay relaxed and avoid overextending fingers or clenching too tightly, as this may affect calibration results and control accuracy.

4.2 gesture_ctrled_rohand

cd /path/to/demo/gesture_ctrled_rohand
pip install -r requirements.txt

After installation is complete, enter 

python gesture_ctrled_hand.py

Can start the demo

Important Notes:

  • You need to select single-dimensional force or three-dimensional force based on the dexterous hand model you purchased. If you purchased AP001, choose TACS_DOT_MATRIX; otherwise, select TACS_3D_FORCE, then choose left or right hand.

  • The vision demo can only recognize gestures with the palm facing the camera and fingers pointing upward. Other postures may result in recognition inaccuracies.

4.3 gForce_ctrled_rohand

cd /path/to/demo/gForce_ctrled_rohand
pip install -r requirements.txt

After installation is complete, enter 

python gForce_ctrled_hand.py

Can start the demo

Note: This demo requires pairing with the gForce armband APP. Gestures must be trained in the APP before use.

Demo Operation Tutorial (Ubuntu System)

Ubuntu usually comes with Python pre-installed, no need to download separately.

1. Uninstall Brltty

Ubuntu does not require installation of 341 driver

However, brltty needs to be uninstalled as this dependency can cause abnormal serial port occupation 

sudo apt remove brltty

2. Check If The Dexterous Hand Is Recognized And Add Permissions

After inserting the dexterous hand 

ls /dev/ttyUSB*

Console prints /dev/ttyUSB0

Grant permissions to the serial port 

sudo chmod o+rw /dev/ttyUSB0

3. Create Virtual Environment (optional)

Using a virtual environment ensures that installing demo dependencies does not affect other environments. 

pip3 install virtualenv

# Select any location where you want to configure the virtual environment, "name" is the name of the virtual environment and can be freely modified
cd path/to/your/folder

# Create a virtual environment, "name" is the virtual environment name and can be customized
virtualenv name

# Activate virtual environment
source name/bin/activate

4. Install Required Dependencies

4.1 glove_ctrled_rohand

Since Ubuntu comes with the bleak-winrt library by default, there's no need to install Visual Studio components. 

cd path/to/demo/glove_ctrled_rohand
pip3 install -r requirements.txt

After installation, you need to go to line 269 in glove_ctrled_hand.py 

client = ModbusSerialClient(self.find_comport("CH340") or self.find_comport("USB"), FramerType.RTU, 115200)
if not client.connect():
    print("Failed to connect to Modbus device")
    exit(-1)

Replace self.find_comport("CH340") to "/dev/ttyUSB0" 

client = ModbusSerialClient("/dev/ttyUSB0" or self.find_comport("USB"), FramerType.RTU, 115200)
if not client.connect():
    print("Failed to connect to Modbus device")
    exit(-1)

If you are using the wired version of the glove, after inserting the glove and starting up, please check

ls /dev/ttyACM*

Has it been printed /dev/ttyACM0

If present, please add permissions for this serial port 

sudo chmod o+rw /dev/ttyACM0

After completion, enter： 

python3 glove_ctrled_hand.py

Can start the demo.

Important Notes:

  • You need to select single-dimensional force or three-dimensional force based on the dexterous hand model you purchased. If you purchased AP001, choose TACS_DOT_MATRIX; otherwise, select TACS_3D_FORCE, then choose left or right hand.

  • The glove requires calibration before use. After running the program, repeatedly perform hand opening-closing-opening movements until calibration is complete. During calibration, try to stay relaxed and avoid overextending fingers or clenching too tightly, as this may affect calibration results and control accuracy.

4.2 gesture_ctrled_rohand

cd /path/to/demo/gesture_ctrled_rohand
pip install -r requirements.txt

After installation, the serial port also needs to be modified. 

client = ModbusSerialClient(find_comport("CH340"), FramerType.RTU, 115200)

At line 375, change find_comport("CH340") to "/dev/ttyUSB0" 

client = ModbusSerialClient("/dev/ttyUSB0", FramerType.RTU, 115200)

At line 375, change find_comport("CH340") to "/dev/ttyUSB0": 

python gesture_ctrled_hand.py

Can start the demo.

Important Notes:

  • You need to select single-dimensional force or three-dimensional force based on the dexterous hand model you purchased. If you purchased AP001, choose TACS_DOT_MATRIX; otherwise, select TACS_3D_FORCE, then choose left or right hand.

  • The vision demo can only recognize gestures with the palm facing the camera and fingers pointing upward. Other postures may result in recognition inaccuracies.

4.3 gForce_ctrled_rohand

cd /path/to/demo/gForce_ctrled_rohand
pip install -r requirements.txt

Similarly modify the serial port 

client = ModbusSerialClient(find_comport("CH340") or find_comport("USB"), FramerType.RTU, 115200)

At line 122, change find_comport("CH340") to "/dev/ttyUSB0" 

client = ModbusSerialClient("/dev/ttyUSB0" or find_comport("USB"), FramerType.RTU, 115200)

After completion, enter 

python gForce_ctrled_hand.py

Can start the demo

Note: This demo requires pairing with the gForce armband APP. Gestures must be trained in the APP before use.

Demo Run Example

1. gesture_ctrled_rohand

gif1

2. glove_ctrled_rohand

gif2

3. gForce_ctrled_rohand

gif3