ROS Integration

The DexHand package provides comprehensive ROS integration supporting both ROS1 and ROS2 environments through a unified API.

Setup and Launch

Start the ROS node with the default configuration:

# Launch the ROS node
python examples/ros_node/dexhand_ros.py

# For testing, run the demo publisher
python examples/ros_node/dexhand_ros_publisher_demo.py --hands right --cycle-time 3.0

# Or run the continuous motion publisher
python examples/ros_node/continuous_joint_publisher.py --pattern sine --amplitude 30 --period 5

Configuration Options

The ROS node behavior can be customized through config/config.yaml under the ROS_Node section:

Topic and Service Interface

Topics

Topic (default)	Type	Direction	Description
/left_hand/joint_commands	sensor_msgs/JointState	Input	Left hand joint commands
/right_hand/joint_commands	sensor_msgs/JointState	Input	Right hand joint commands
/left_hand/joint_states	sensor_msgs/JointState	Output	Left hand joint feedback
/right_hand/joint_states	sensor_msgs/JointState	Output	Right hand joint feedback
/left_hand/touch_sensors	Float64MultiArray	Output	Left hand touch sensor data
/right_hand/touch_sensors	Float64MultiArray	Output	Right hand touch sensor data
/left_hand/motor_feedback	Float64MultiArray	Output	Left hand detailed motor data
/right_hand/motor_feedback	Float64MultiArray	Output	Right hand detailed motor data

Services

Feedback Data Structure

Touch Sensor Data

Touch sensor data is published as Float64MultiArray with 40 values (8 values for each of the 5 fingers):

Format per finger: [timestamp, normal_force, normal_force_delta, tangential_force, tangential_force_delta, direction, proximity, temperature]
Direction is in degrees (0-359, fingertip is 0°) or -1 if invalid
Finger mapping: * Index 0: thumb (th) * Index 1: index finger (ff) * Index 2: middle finger (mf) * Index 3: ring finger (rf) * Index 4: little finger (lf)

Motor Feedback Data

Motor feedback is published as Float64MultiArray with 84 values (7 values for each of the 12 motors):

Format per motor: [timestamp, angle, encoder_position, current, velocity, error_code, impedance]
Impedance values: lower values indicate higher resistance to movement
Motor ordering follows joint names: th_dip, th_mcp, th_rot, ff_spr, etc.

Example Usage

Publishing Joint Commands

# Simple ROS1/ROS2 agnostic publisher
from ros_compat import create_ros_node, Publisher
from sensor_msgs.msg import JointState

# Create node and publisher
node = create_ros_node('hand_command_publisher')
pub = Publisher('/right_hand/joint_commands', JointState, queue_size=10)

# Create message
msg = JointState()
msg.name = ['r_f_joint1_2', 'r_f_joint2_2']  # URDF joint names
msg.position = [0.5, 0.5]  # Radians

# Publish
msg.header.stamp = node.get_time()
pub.publish(msg)