Lidar Driver Config

name: lidar-driver-config description: 'Configure lidar drivers for ROS 2. Use when setting up LD19, RPLidar, Hokuyo, or Velodyne driver parameters.'

LIDAR Driver Configuration for ROS 2

LIDAR Options for Mobile Robots

LDROBOT LD19 Configuration

The LD19 is a common budget 2D lidar. Uses a CP2102 USB-serial adapter at 230400 baud.

Launch File

from launch import LaunchDescription
from launch_ros.actions import Node

def generate_launch_description():
    return LaunchDescription([
        Node(
            package='ldlidar_stl_ros2',
            executable='ldlidar_stl_ros2_node',
            name='ldlidar',
            output='screen',
            parameters=[{
                'product_name': 'LDLiDAR_LD19',
                'topic_name': '/scan',
                'frame_id': 'laser_frame',
                'port_name': '/dev/lidar',
                'port_baudrate': 230400,
                'laser_scan_dir': True,          # True = counterclockwise (ROS standard)
                'enable_angle_crop_func': False,
                'angle_crop_min': 0.0,
                'angle_crop_max': 360.0,
            }],
        ),
    ])

Key Parameters

• laser_scan_dir: The LD19 scans clockwise physically. Set True to flip to counterclockwise (ROS convention). If your map builds mirrored, toggle this.
• frame_id: Must match the URDF link name where the lidar is mounted. This connects the scan data to the robot's TF tree.
• enable_angle_crop_func: Set True to exclude angular ranges where the robot's body is visible. Useful when the lidar is partially obstructed by the chassis.

RPLidar / Slamtec Configuration

rplidar_ros (Legacy Package)

Node(
    package='rplidar_ros',
    executable='rplidar_node',
    name='rplidar',
    parameters=[{
        'serial_port': '/dev/rplidar',
        'serial_baudrate': 256000,         # A2: 256000, A1: 115200
        'frame_id': 'laser_frame',
        'inverted': False,
        'angle_compensate': True,
        'scan_mode': 'Standard',           # or 'Express', 'Boost', 'Sensitivity'
    }],
)

sllidar_ros2 (Current Slamtec Package)

Node(
    package='sllidar_ros2',
    executable='sllidar_node',
    name='sllidar',
    parameters=[{
        'channel_type': 'serial',
        'serial_port': '/dev/rplidar',
        'serial_baudrate': 256000,
        'frame_id': 'laser_frame',
        'inverted': False,
        'angle_compensate': True,
        'scan_mode': 'Standard',
        'scan_frequency': 10.0,            # Hz, adjustable 5-15 Hz
    }],
)

Scan Modes

Hokuyo URG Configuration

Node(
    package='urg_node2',
    executable='urg_node2_node',
    name='urg_node',
    parameters=[{
        'ip_address': '',                  # empty for USB
        'serial_port': '/dev/hokuyo',
        'serial_baud': 115200,
        'frame_id': 'laser_frame',
        'angle_min': -2.356,               # -135° in radians
        'angle_max': 2.356,                # +135° in radians
        'range_min': 0.02,
        'range_max': 5.6,
        'cluster': 1,                      # group N adjacent points
        'skip': 0,                         # skip N scans between publishes
    }],
)

Frame ID and URDF Integration

The frame_id parameter connects laser data to the robot model. In URDF:

<link name="laser_frame">
  <visual>
    <geometry><cylinder length="0.04" radius="0.035"/></geometry>
  </visual>
</link>

<joint name="laser_joint" type="fixed">
  <parent link="base_link"/>
  <child link="laser_frame"/>
  <origin xyz="0.15 0.0 0.12" rpy="0 0 0"/>
</joint>

The xyz values define the lidar's position relative to base_link. Measure carefully — a 2cm error in the lidar mount position causes SLAM map distortion.

Inverted Mounting

If the lidar is mounted upside-down (common for under-chassis mounting):

• Set inverted: True in the driver — this flips the scan vertically
• In URDF, add rpy="3.14159 0 0" (180° roll) to the joint origin
• Verify with RViz: scan points should appear at the correct height and orientation

Orientation and Scan Direction

ROS sensor_msgs/LaserScan convention:

• angle_min to angle_max sweeps counterclockwise when viewed from above
• 0° points forward (along the robot's x-axis)

If your SLAM map appears mirrored, the scan direction is wrong. Solutions:

1. Driver parameter: laser_scan_dir, inverted, or scan_dir_reverse
2. URDF: Rotate the laser frame by 180° around z: rpy="0 0 3.14159"

Angle Offset

If the lidar's zero-angle doesn't align with the robot's forward direction:

# Rotate the scan by adding angle_offset in the URDF transform
# If lidar zero is 90° to the right of robot forward:
<origin xyz="0.15 0.0 0.12" rpy="0 0 1.5708"/>  # 90° yaw

Alternatively, some drivers have an angle_offset parameter. Either approach works; URDF is preferred because it's visible in RViz and doesn't depend on driver support.

Multi-LIDAR Setup

For robots with multiple lidars (e.g., front + rear for full 360° coverage):

Node(
    package='ldlidar_stl_ros2',
    executable='ldlidar_stl_ros2_node',
    name='lidar_front',
    parameters=[{
        'port_name': '/dev/lidar_front',
        'topic_name': '/scan_front',
        'frame_id': 'laser_front_frame',
    }],
),
Node(
    package='ldlidar_stl_ros2',
    executable='ldlidar_stl_ros2_node',
    name='lidar_rear',
    parameters=[{
        'port_name': '/dev/lidar_rear',
        'topic_name': '/scan_rear',
        'frame_id': 'laser_rear_frame',
    }],
),

Merging Scans

Use ira_laser_tools to combine multiple LaserScan topics:

Node(
    package='ira_laser_tools',
    executable='laserscan_multi_merger',
    name='scan_merger',
    parameters=[{
        'destination_frame': 'base_link',
        'cloud_destination_topic': '/merged_cloud',
        'scan_destination_topic': '/scan',
        'laserscan_topics': '/scan_front /scan_rear',
        'angle_min': -3.14159,
        'angle_max': 3.14159,
        'range_min': 0.05,
        'range_max': 12.0,
    }],
)

The merger transforms each scan into base_link frame (using TF), combines them, and publishes a unified /scan topic for SLAM and navigation.

Laser Filtering

Use laser_filters to clean up scan data before feeding it to SLAM:

scan_to_scan_filter_chain:
  ros__parameters:
    filter1:
      name: range
      type: laser_filters/LaserScanRangeFilter
      params:
        lower_threshold: 0.05
        upper_threshold: 10.0
    filter2:
      name: angular
      type: laser_filters/LaserScanAngularBoundsFilter
      params:
        lower_angle: -2.5
        upper_angle: 2.5
    filter3:
      name: median
      type: laser_filters/LaserScanMedianFilter
      params:
        window_size: 3

Filters remove out-of-range readings, crop angular sections (exclude robot body), and smooth noisy measurements.