Skip to main content

Sensor Simulation Quickstart Guide

Prerequisites​

Before starting with sensor simulation, ensure you have:

  • ROS 2: Humble Hawksbill or later installed and sourced
  • Gazebo Garden: Or compatible version with sensor plugins
  • Basic ROS 2 Knowledge: Understanding of sensor message types
  • Gazebo Simulation: Working physics simulation environment
  • System Requirements: Adequate processing power for sensor simulation

Overview of Sensor Types​

This guide covers three primary sensor types commonly used in robotics:

  1. LiDAR Sensors: Generate 3D point clouds for mapping and obstacle detection
  2. Depth Cameras: Provide 2.5D information combining color and depth data
  3. IMU Sensors: Measure orientation and acceleration for navigation

LiDAR Sensor Setup​

1. Basic LiDAR Configuration​

Create a simple LiDAR sensor configuration in your robot's SDF file:

<sensor name="lidar_sensor" type="lidar">
  <pose>0.5 0 0.3 0 0 0</pose>
  <topic>lidar_scan</topic>
  <update_rate>10</update_rate>
  <ray>
    <scan>
      <horizontal>
        <samples>360</samples>
        <resolution>1</resolution>
        <min_angle>-3.14159</min_angle>
        <max_angle>3.14159</max_angle>
      </horizontal>
    </scan>
    <range>
      <min>0.1</min>
      <max>30.0</max>
      <resolution>0.01</resolution>
    </range>
    <noise>
      <type>gaussian</type>
      <mean>0.0</mean>
      <stddev>0.01</stddev>
    </noise>
  </ray>
</sensor>

2. Launching LiDAR Simulation​

To launch a simulation with LiDAR:

# Launch your simulation with LiDAR sensor
gz sim -r your_world_with_lidar.sdf

# Monitor LiDAR data in another terminal
source /opt/ros/humble/setup.bash
ros2 topic echo /lidar_scan sensor_msgs/msg/LaserScan

3. LiDAR Data Analysis​

LiDAR data is published as sensor_msgs/LaserScan messages containing:

  • ranges: Array of distance measurements
  • intensities: Optional intensity values
  • angle_min, angle_max: Angular range of the scan
  • angle_increment: Angular resolution
  • time_increment: Time between measurements
  • scan_time: Time between scans
  • range_min, range_max: Valid range of measurements

Depth Camera Setup​

1. Basic Depth Camera Configuration​

Add a depth camera to your robot:

<sensor name="depth_camera" type="depth_camera">
  <pose>0.1 0 0.2 0 0 0</pose>
  <topic>depth_camera/depth/image_raw</topic>
  <camera>
    <horizontal_fov>1.047</horizontal_fov>
    <image>
      <width>640</width>
      <height>480</height>
    </image>
    <clip>
      <near>0.1</near>
      <far>10</far>
    </clip>
  </camera>
  <noise>
    <type>gaussian</type>
    <mean>0.0</mean>
    <stddev>0.01</stddev>
  </noise>
</sensor>

2. Launching Depth Camera Simulation​

# Launch simulation with depth camera
gz sim -r your_world_with_camera.sdf

# Monitor depth camera data
source /opt/ros/humble/setup.bash
ros2 topic echo /depth_camera/depth/image_raw sensor_msgs/msg/Image
ros2 topic echo /depth_camera/color/image_raw sensor_msgs/msg/Image

3. Depth Camera Data Analysis​

Depth camera provides multiple data streams:

  • depth/image_raw: Depth information (16-bit float)
  • color/image_raw: Color image data (8-bit RGB)
  • camera_info: Camera calibration parameters

IMU Sensor Setup​

1. Basic IMU Configuration​

Add an IMU sensor to your robot:

<sensor name="imu_sensor" type="imu">
  <pose>0 0 0.1 0 0 0</pose>
  <topic>imu/data</topic>
  <update_rate>100</update_rate>
  <imu>
    <angular_velocity>
      <x>
        <noise>
          <type>gaussian</type>
          <mean>0.0</mean>
          <stddev>0.001</stddev>
        </noise>
      </x>
      <y>
        <noise>
          <type>gaussian</type>
          <mean>0.0</mean>
          <stddev>0.001</stddev>
        </noise>
      </y>
      <z>
        <noise>
          <type>gaussian</type>
          <mean>0.0</mean>
          <stddev>0.001</stddev>
        </noise>
      </z>
    </angular_velocity>
    <linear_acceleration>
      <x>
        <noise>
          <type>gaussian</type>
          <mean>0.0</mean>
          <stddev>0.017</stddev>
        </noise>
      </x>
      <y>
        <noise>
          <type>gaussian</type>
          <mean>0.0</mean>
          <stddev>0.017</stddev>
        </noise>
      </y>
      <z>
        <noise>
          <type>gaussian</type>
          <mean>0.0</mean>
          <stddev>0.017</stddev>
        </noise>
      </z>
    </linear_acceleration>
  </imu>
</sensor>

2. Launching IMU Simulation​

# Launch simulation with IMU
gz sim -r your_world_with_imu.sdf

# Monitor IMU data
source /opt/ros/humble/setup.bash
ros2 topic echo /imu/data sensor_msgs/msg/Imu

3. IMU Data Analysis​

IMU data is published as sensor_msgs/Imu messages containing:

  • orientation: Quaternion representing orientation
  • angular_velocity: Angular velocity in x, y, z
  • linear_acceleration: Linear acceleration in x, y, z
  • Covariance matrices for uncertainty information

Sensor Integration in Simulation​

1. Complete Robot with Multiple Sensors​

Here's an example of a robot with all three sensor types:

<model name="robot_with_sensors">
  <link name="base_link">
    <!-- Robot chassis -->
    <collision name="collision">
      <geometry>
        <box><size>0.5 0.3 0.3</size></box>
      </geometry>
    </collision>
    <visual name="visual">
      <geometry>
        <box><size>0.5 0.3 0.3</size></box>
      </geometry>
    </visual>
    <inertial>
      <mass>1.0</mass>
      <inertia><ixx>0.01</ixx><ixy>0</ixy><ixz>0</ixz><iyy>0.01</iyy><iyz>0</iyz><izz>0.01</izz></inertia>
    </inertial>
  </link>

  <!-- LiDAR sensor -->
  <sensor name="lidar" type="lidar">
    <pose>0.2 0 0.4 0 0 0</pose>
    <topic>lidar_scan</topic>
    <update_rate>10</update_rate>
    <ray>
      <scan><horizontal><samples>360</samples><resolution>1</resolution><min_angle>-3.14</min_angle><max_angle>3.14</max_angle></horizontal></scan>
      <range><min>0.1</min><max>10.0</max><resolution>0.01</resolution></range>
    </ray>
  </sensor>

  <!-- Depth camera -->
  <sensor name="camera" type="depth_camera">
    <pose>0.1 0 0.3 0 0 0</pose>
    <topic>depth_camera</topic>
    <camera><horizontal_fov>1.047</horizontal_fov><image><width>640</width><height>480</height></image><clip><near>0.1</near><far>10</far></clip></camera>
  </sensor>

  <!-- IMU sensor -->
  <sensor name="imu" type="imu">
    <pose>0 0 0.2 0 0 0</pose>
    <topic>imu/data</topic>
    <update_rate>100</update_rate>
  </sensor>
</model>

2. Monitoring All Sensors​

Monitor all sensor data simultaneously:

# Terminal 1: Launch simulation
gz sim -r robot_with_sensors.sdf

# Terminal 2: Monitor LiDAR
source /opt/ros/humble/setup.bash
ros2 topic echo /lidar_scan sensor_msgs/msg/LaserScan

# Terminal 3: Monitor camera
ros2 topic echo /depth_camera/depth/image_raw sensor_msgs/msg/Image

# Terminal 4: Monitor IMU
ros2 topic echo /imu/data sensor_msgs/msg/Imu

Validation and Testing​

1. Basic Validation Tests​

Verify each sensor is working properly:

LiDAR Validation​

# Check if topic exists and has data
ros2 topic list | grep lidar
ros2 topic info /lidar_scan

# Check data rate
ros2 topic hz /lidar_scan

Camera Validation​

# Check image dimensions and format
ros2 topic echo /depth_camera/color/image_raw --field width
ros2 topic echo /depth_camera/color/image_raw --field height

IMU Validation​

# Check if values are reasonable (around 9.8 for gravity in z-axis)
ros2 topic echo /imu/data --field linear_acceleration.z

2. Visualization Tools​

Use RViz2 to visualize sensor data:

# Launch RViz2
source /opt/ros/humble/setup.bash
rviz2

# Add displays for:
# - LaserScan for LiDAR data
# - Image for camera data
# - Imu for IMU data
# - RobotModel to see robot state

Common Issues and Solutions​

Issue: No sensor data being published​

Solution:

  • Check sensor configuration in SDF/URDF
  • Verify Gazebo simulation is running
  • Check topic names match between configuration and monitoring

Issue: Unrealistic sensor values​

Solution:

  • Verify sensor mounting position and orientation
  • Check noise parameters and range limits
  • Ensure physics simulation is active

Issue: High CPU usage with sensors​

Solution:

  • Reduce update rates for less critical sensors
  • Lower resolution parameters
  • Simplify environment geometry

Issue: Sensor data lag or delay​

Solution:

  • Check real-time factor in physics configuration
  • Optimize simulation step size
  • Verify network performance if using distributed simulation

Best Practices​

  1. Realistic Parameters: Use sensor parameters that match real hardware specifications
  2. Appropriate Update Rates: Balance between data quality and performance
  3. Noise Modeling: Include realistic noise models for authentic behavior
  4. Coordinate Frames: Ensure consistent coordinate systems across sensors
  5. Validation: Regularly validate sensor outputs against expected ranges

Next Steps​

After completing this quickstart guide, you should be able to:

  • Configure LiDAR, depth camera, and IMU sensors in Gazebo
  • Launch simulations with multiple sensor types
  • Monitor and validate sensor data streams
  • Integrate sensors into a complete robot model

Continue to the next section to learn about advanced sensor fusion techniques and data processing.