BrainCo Hand ROS2 Project

English | 简体中文

Project Overview

BrainCo Hand ROS2 is a complete ROS 2 software package collection that provides a full solution for BrainCo Revo2 dexterous hand, from hardware drivers to simulation and motion planning. This project is developed based on ROS 2 Humble, supporting single-hand and dual-hand configurations, providing hardware control, Gazebo simulation, MoveIt motion planning, and other features.

Key Features

• Complete Hardware Driver Support: Supports Modbus, CAN FD, and EtherCAT communication protocols
• ros2_control Integration: Complete ros2_control hardware interface implementation
• Gazebo Simulation Environment: High-fidelity physics simulation based on Ignition Gazebo 6
• MoveIt Motion Planning: Complete MoveIt 2 integration supporting advanced motion planning
• Real-time Control: High-frequency control loop for precise finger manipulation
• Dual-hand Support: Supports left hand, right hand, and simultaneous dual-hand control
• Robot Arm Integration: Provides RM65 robot arm integration demo with Revo2 dexterous hand

System Requirements

Basic Environment

• Operating System: Ubuntu 22.04
• ROS Version: ROS 2 Humble
• Python Version: Python 3.8+

Hardware Requirements (Hardware Control Mode)

Modbus Mode (Default)

• Modbus serial device (e.g., /dev/ttyUSB0)
• Serial port permission configuration (user must be in dialout group)

CAN FD Mode (Optional)

• ZLG USB-CAN FD device (e.g., USBCANFD-200U)
• CAN FD bus connection

Simulation Environment Requirements

• Gazebo: Ignition Gazebo 6 (for simulation packages)
• MoveIt 2: For motion planning packages

Project Structure

This project contains the following main packages:

brainco_hand_ros2/
├── revo2_description/                    # Revo2 dexterous hand URDF description package
├── brainco_hardware/                     # Hardware driver packages
│   ├── brainco_hand_driver/            # Revo2 dexterous hand hardware driver
│   └── stark_ethercat/brainco_hand_ethercat_driver/                  # EtherCAT driver
├── brainco_gazebo/                      # Gazebo simulation package
├── brainco_moveit_config/               # MoveIt configuration package
└── revo2_with_rm65_demo/                # RM65 robot arm integration demo
    ├── gazebo_rm_65_6f_with_revo2_demo/ # RM65+Revo2 Gazebo simulation
    └── rm65_with_revo2_right_moveit_config/ # RM65+Revo2 MoveIt configuration

Package Descriptions

1. revo2_description

Function: URDF model description package for Revo2 dexterous hand, providing 3D visualization of the robot model.

Key Features:

• Left and right hand URDF models
• RViz visualization support
• Complete joint and link definitions

Detailed Documentation: Please refer to revo2_description/README_CN.md

2. brainco_hardware

Function: Hardware driver package for Revo2 dexterous hand, providing ros2_control-based hardware interface.

Key Features:

• Supports Modbus, CAN FD, and EtherCAT communication protocols
• Complete ros2_control hardware interface implementation
• Supports single-hand and dual-hand configurations
• MoveIt integration support
• Real-time joint position and velocity feedback

Detailed Documentation: Please refer to brainco_hardware/brainco_hand_driver/README_CN.md

EtherCAT Detailed Documentation: Please refer to brainco_hardware/stark_ethercat/brainco_hand_ethercat_driver/README_CN.md

3. brainco_gazebo

Function: Gazebo simulation package for Revo2 dexterous hand, providing complete physics simulation environment.

Key Features:

• Physics simulation based on Ignition Gazebo 6
• Supports single-hand and dual-hand simulation
• Gazebo-MoveIt integration
• RViz visualization integration

Detailed Documentation: Please refer to brainco_gazebo/README_CN.md

4. brainco_moveit_config

Function: MoveIt configuration package for Revo2 dexterous hand, providing motion planning functionality.

Key Features:

• Supports left hand, right hand, and dual-hand configurations
• Complete MoveIt 2 integration
• FakeSystem support

Detailed Documentation: Please refer to brainco_moveit_config/README_CN.md

5. revo2_with_rm65_demo

Function: Integration demo package for RM65 robot arm with Revo2 dexterous hand.

Sub-packages:

5.1 gazebo_rm_65_6f_with_revo2_demo

Function: Gazebo simulation for RM65 robot arm with Revo2 dexterous hand.

Detailed Documentation: Please refer to revo2_with_rm65_demo/gazebo_rm_65_6f_with_revo2_demo/README_CN.md

5.2 rm65_with_revo2_right_moveit_config

Function: MoveIt configuration for RM65 robot arm with Revo2 right hand.

Detailed Documentation: Please refer to revo2_with_rm65_demo/rm65_with_revo2_right_moveit_config/README_CN.md

Quick Start

1. Environment Setup

# Check ROS 2 environment
echo $ROS_DISTRO  # Should output: humble

# Create workspace (if not already created)
mkdir -p <workspace>/src
cd <workspace>

Note: Please replace <workspace> with your workspace path, for example ~/brainco_ws.

2. Clone Repository

# Enter workspace src directory
cd <workspace>/src

# Clone main repository
git clone https://github.com/BrainCoTech/brainco_hand_ros2.git
cd brainco_hand_ros2

# Use vcs tool to clone dependency repositories
vcs import . < brainco_hand.repos --recursive --skip-existing

Note: If your system does not have the vcs tool installed, please install it first:

sudo apt-get install python3-vcstool

3. Install Dependencies and Build

# Return to workspace root directory
cd <workspace>

# Update package list
sudo apt-get update

# Update rosdep database
rosdep update

# Install all dependencies
rosdep install --from-paths src --ignore-src --rosdistro $ROS_DISTRO -y

# Build all packages
colcon build --symlink-install --cmake-args -DCMAKE_BUILD_TYPE=Release --continue-on-error

# Source workspace
source install/setup.bash

Usage Scenarios

Scenario 1: Hardware Control (Real Dexterous Hand)

If you have real Revo2 dexterous hand hardware, you can use brainco_hand_driver for control:

# Launch right hand system (Modbus mode)
ros2 launch brainco_hand_driver revo2_system.launch.py hand_type:=right

# Launch right hand system (with MoveIt)
ros2 launch brainco_hand_driver revo2_real_moveit.launch.py hand_type:=right

# Launch dual-hand system
ros2 launch brainco_hand_driver dual_revo2_real_moveit.launch.py

Detailed Instructions: Please refer to brainco_hardware/brainco_hand_driver/README_CN.md

Scenario 2: Gazebo Simulation

If you want to develop and test without hardware, you can use Gazebo simulation:

# Launch single-hand simulation
ros2 launch brainco_gazebo revo2_hand_gazebo.launch.py hand_type:=right

# Launch dual-hand simulation
ros2 launch brainco_gazebo dual_revo2_hand_gazebo.launch.py

# Launch simulation with MoveIt
ros2 launch brainco_gazebo revo2_hand_gazebo_moveit.launch.py hand_type:=right

Detailed Instructions: Please refer to brainco_gazebo/README_CN.md

Scenario 3: MoveIt Motion Planning (No Hardware)

If you want to use MoveIt for motion planning but don't need hardware or simulation, you can use FakeSystem:

# Launch right hand MoveIt (FakeSystem)
ros2 launch brainco_moveit_config revo2_right_moveit.launch.py

# Launch dual-hand MoveIt (FakeSystem)
ros2 launch brainco_moveit_config dual_revo2_moveit.launch.py

Detailed Instructions: Please refer to brainco_moveit_config/README_CN.md

Scenario 4: Robot Arm Integration Demo (FakeSystem)

If you want to test the integration of RM65 robot arm with Revo2 dexterous hand:

# Launch Gazebo simulation
ros2 launch gazebo_rm_65_6f_with_revo2_demo gazebo_rm_65_6f_with_revo2.launch.py

# Launch MoveIt configuration (FakeSystem)
ros2 launch rm65_with_revo2_right_moveit_config rm65_with_revo2_right_moveit.launch.py

Detailed Instructions:

• revo2_with_rm65_demo/gazebo_rm_65_6f_with_revo2_demo/README_CN.md
• revo2_with_rm65_demo/rm65_with_revo2_right_moveit_config/README_CN.md

Dual-hand MoveIt Simulation Demo

Control Interfaces

Topic Interfaces

All packages provide standard ROS 2 Topic interfaces:

• Joint States: /joint_states (sensor_msgs/JointState)
• Trajectory Commands: /xxx_revo2_hand_controller/joint_trajectory (trajectory_msgs/JointTrajectory)

Action Interfaces

• Trajectory Execution: /xxx_revo2_hand_controller/follow_joint_trajectory (control_msgs/action/FollowJointTrajectory)

Control Examples

Right Hand Open Palm

ros2 topic pub --once /right_revo2_hand_controller/joint_trajectory \
  trajectory_msgs/msg/JointTrajectory \
  '{
    joint_names: [
      "right_thumb_proximal_joint",
      "right_thumb_metacarpal_joint",
      "right_index_proximal_joint",
      "right_middle_proximal_joint",
      "right_ring_proximal_joint",
      "right_pinky_proximal_joint"
    ],
    points: [{
      positions: [0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
      time_from_start: {sec: 1}
    }]
  }'

Right Hand Make Fist

ros2 topic pub --once /right_revo2_hand_controller/joint_trajectory \
  trajectory_msgs/msg/JointTrajectory \
  '{
    joint_names: [
      "right_thumb_proximal_joint",
      "right_thumb_metacarpal_joint",
      "right_index_proximal_joint",
      "right_middle_proximal_joint",
      "right_ring_proximal_joint",
      "right_pinky_proximal_joint"
    ],
    points: [{
      positions: [0.1, 0.8, 1.4, 1.4, 1.4, 1.4],
      time_from_start: {sec: 2}
    }]
  }'

Monitoring and Debugging

System Status Check

# List all running nodes
ros2 node list

# List all controllers
ros2 control list_controllers

# Check hardware components
ros2 control list_hardware_components

# Check hardware interfaces
ros2 control list_hardware_interfaces

# List all topics
ros2 topic list

# List all actions
ros2 action list

# Monitor joint states
ros2 topic echo /joint_states

Contact

For questions or suggestions, please contact the development team.