Skip to main content

​
Overview

Gazebo is a 3D robot simulator that provides realistic physics, sensor simulation, and visualization. It integrates with ROS2 to test robot behavior before deploying to hardware, saving time and preventing damage to physical robots.

Physics Simulation

Realistic dynamics, collisions, and friction

Sensor Simulation

LiDAR, cameras, IMU, and other sensors
Gazebo simulation is optional for this course. Focus on hardware implementation first. Use simulation if you have extra time or want to test navigation algorithms safely.

​
Installation

​
Gazebo Harmonic (for ROS2 Jazzy)

# Install Gazebo Harmonic
sudo apt update
sudo apt install ros-jazzy-ros-gz

# Install Gazebo itself
sudo apt install gz-harmonic

# Verify installation
gz sim --version
# Output: Gazebo Sim, version 8.x.x

​
URDF with Gazebo Plugins

​
Adding Gazebo-Specific Properties

mecanum_robot_gazebo.urdf.xacro
<?xml version="1.0"?>
<robot xmlns:xacro="http://www.ros.org/wiki/xacro" name="mecanum_robot">
  <!-- Include base robot URDF -->
  <xacro:include filename="$(find mecanum_description)/urdf/mecanum_robot.urdf.xacro"/>

  <!-- Gazebo-specific properties for chassis -->
  <gazebo reference="chassis">
    <material>Gazebo/Blue</material>
    <mu1>0.2</mu1>  <!-- Friction coefficient -->
    <mu2>0.2</mu2>
  </gazebo>

  <!-- Gazebo properties for wheels -->
  <xacro:macro name="wheel_gazebo" params="prefix">
    <gazebo reference="${prefix}_wheel">
      <material>Gazebo/Black</material>
      <mu1>1.0</mu1>  <!-- High friction for traction -->
      <mu2>1.0</mu2>
      <kp>1000000.0</kp>  <!-- Contact stiffness -->
      <kd>100.0</kd>      <!-- Contact damping -->
    </gazebo>
  </xacro:macro>

  <xacro:wheel_gazebo prefix="front_left"/>
  <xacro:wheel_gazebo prefix="front_right"/>
  <xacro:wheel_gazebo prefix="rear_left"/>
  <xacro:wheel_gazebo prefix="rear_right"/>

  <!-- LiDAR sensor plugin -->
  <gazebo reference="lidar_link">
    <sensor name="lidar" type="gpu_lidar">
      <pose>0 0 0 0 0 0</pose>
      <topic>/scan</topic>
      <update_rate>5.5</update_rate>  <!-- RPLIDAR A1M8: 5.5 Hz -->
      <lidar>
        <scan>
          <horizontal>
            <samples>360</samples>
            <resolution>1.0</resolution>
            <min_angle>-3.14159</min_angle>
            <max_angle>3.14159</max_angle>
          </horizontal>
        </scan>
        <range>
          <min>0.15</min>  <!-- RPLIDAR A1M8: 0.15m min -->
          <max>12.0</max>  <!-- RPLIDAR A1M8: 12m max -->
          <resolution>0.01</resolution>
        </range>
      </lidar>
      <always_on>true</always_on>
      <visualize>true</visualize>
    </sensor>
  </gazebo>

  <!-- IMU sensor plugin -->
  <gazebo reference="imu_link">
    <sensor name="imu" type="imu">
      <always_on>true</always_on>
      <update_rate>50</update_rate>  <!-- 50 Hz -->
      <topic>/imu/data</topic>
      <imu>
        <angular_velocity>
          <x>
            <noise type="gaussian">
              <mean>0.0</mean>
              <stddev>0.0002</stddev>
            </noise>
          </x>
          <!-- y, z similar -->
        </angular_velocity>
        <linear_acceleration>
          <x>
            <noise type="gaussian">
              <mean>0.0</mean>
              <stddev>0.001</stddev>
            </noise>
          </x>
          <!-- y, z similar -->
        </linear_acceleration>
      </imu>
    </sensor>
  </gazebo>

  <!-- Differential drive controller (approximation for mecanum) -->
  <gazebo>
    <plugin filename="libignition-gazebo-diff-drive-system.so"
            name="ignition::gazebo::systems::DiffDrive">
      <left_joint>front_left_wheel_joint</left_joint>
      <left_joint>rear_left_wheel_joint</left_joint>
      <right_joint>front_right_wheel_joint</right_joint>
      <right_joint>rear_right_wheel_joint</right_joint>
      <wheel_separation>0.20</wheel_separation>
      <wheel_diameter>0.10</wheel_diameter>
      <odom_publish_frequency>20</odom_publish_frequency>
      <topic>/cmd_vel</topic>
      <odom_topic>/odom</odom_topic>
      <tf_topic>/tf</tf_topic>
    </plugin>
  </gazebo>
</robot>
Gazebo doesn’t have a native mecanum drive controller. Options:
  1. Use DiffDrive plugin (approximation, no lateral movement)
  2. Write custom mecanum plugin (advanced)
  3. Use Gazebo + ros2_control for accurate simulation

​
Launch Gazebo Simulation

gazebo.launch.py
import os
from ament_index_python.packages import get_package_share_directory
from launch import LaunchDescription
from launch.actions import IncludeLaunchDescription, ExecuteProcess
from launch.launch_description_sources import PythonLaunchDescriptionSource
from launch_ros.actions import Node
import xacro

def generate_launch_description():
    pkg_share = get_package_share_directory('mecanum_description')
    xacro_file = os.path.join(pkg_share, 'urdf', 'mecanum_robot_gazebo.urdf.xacro')
    world_file = os.path.join(pkg_share, 'worlds', 'empty.world')

    # Process xacro to URDF
    robot_description = xacro.process_file(xacro_file).toxml()

    return LaunchDescription([
        # Launch Gazebo server
        ExecuteProcess(
            cmd=['gz', 'sim', '-r', world_file],
            output='screen'
        ),

        # Spawn robot in Gazebo
        Node(
            package='ros_gz_sim',
            executable='create',
            arguments=[
                '-name', 'mecanum_robot',
                '-topic', '/robot_description',
                '-x', '0',
                '-y', '0',
                '-z', '0.1',
            ],
            output='screen'
        ),

        # Robot state publisher
        Node(
            package='robot_state_publisher',
            executable='robot_state_publisher',
            parameters=[{'robot_description': robot_description}],
        ),

        # Bridge Gazebo topics to ROS2
        Node(
            package='ros_gz_bridge',
            executable='parameter_bridge',
            arguments=[
                '/scan@sensor_msgs/msg/LaserScan[gz.msgs.LaserScan',
                '/imu/data@sensor_msgs/msg/Imu[gz.msgs.IMU',
                '/cmd_vel@geometry_msgs/msg/Twist]gz.msgs.Twist',
                '/odom@nav_msgs/msg/Odometry[gz.msgs.Odometry',
            ],
            output='screen'
        ),
    ])

​
Creating Custom Worlds

worlds/lab_environment.world
<?xml version="1.0"?>
<sdf version="1.9">
  <world name="lab">
    <!-- Physics -->
    <physics name="1ms" type="ignored">
      <max_step_size>0.001</max_step_size>
      <real_time_factor>1.0</real_time_factor>
    </physics>

    <!-- Lighting -->
    <light type="directional" name="sun">
      <cast_shadows>true</cast_shadows>
      <pose>0 0 10 0 0 0</pose>
      <diffuse>0.8 0.8 0.8 1</diffuse>
      <specular>0.2 0.2 0.2 1</specular>
      <direction>-0.5 0.1 -0.9</direction>
    </light>

    <!-- Ground plane -->
    <model name="ground_plane">
      <static>true</static>
      <link name="link">
        <collision name="collision">
          <geometry>
            <plane>
              <normal>0 0 1</normal>
            </plane>
          </geometry>
        </collision>
        <visual name="visual">
          <geometry>
            <plane>
              <normal>0 0 1</normal>
              <size>100 100</size>
            </plane>
          </geometry>
          <material>
            <ambient>0.8 0.8 0.8 1</ambient>
            <diffuse>0.8 0.8 0.8 1</diffuse>
          </material>
        </visual>
      </link>
    </model>

    <!-- Wall obstacle -->
    <model name="wall1">
      <static>true</static>
      <pose>2 0 0.5 0 0 0</pose>
      <link name="link">
        <collision name="collision">
          <geometry>
            <box><size>0.1 4 1</size></box>
          </geometry>
        </collision>
        <visual name="visual">
          <geometry>
            <box><size>0.1 4 1</size></box>
          </geometry>
          <material>
            <ambient>0.7 0.2 0.2 1</ambient>
          </material>
        </visual>
      </link>
    </model>

    <!-- Box obstacle -->
    <model name="box1">
      <pose>1 1 0.25 0 0 0</pose>
      <link name="link">
        <collision name="collision">
          <geometry>
            <box><size>0.5 0.5 0.5</size></box>
          </geometry>
        </collision>
        <visual name="visual">
          <geometry>
            <box><size>0.5 0.5 0.5</size></box>
          </geometry>
          <material>
            <ambient>0.2 0.2 0.7 1</ambient>
          </material>
        </visual>
      </link>
    </model>
  </world>
</sdf>

​
Testing in Simulation

​
1. Launch Simulation

ros2 launch mecanum_description gazebo.launch.py

​
2. Teleop Control

# In separate terminal
ros2 run teleop_twist_keyboard teleop_twist_keyboard --ros-args -r /cmd_vel:=/cmd_vel

​
3. Verify Sensors

# Check LiDAR data
ros2 topic echo /scan

# Check IMU data
ros2 topic echo /imu/data

# Check odometry
ros2 topic echo /odom

​
4. Visualize in RViz

ros2 run rviz2 rviz2

# Add displays:
# - RobotModel
# - LaserScan (topic: /scan)
# - TF
# - Odometry (topic: /odom)

​
Benefits of Simulation

Safe Testing

  • Test dangerous scenarios without risk
  • No hardware damage from crashes
  • Unlimited experiments
  • Reset instantly

Algorithm Development

  • Test navigation without waiting for hardware
  • Iterate quickly on path planning
  • Debug localization issues
  • Tune parameters safely

Reproducibility

  • Same environment every time
  • Controlled test scenarios
  • Benchmark performance
  • Share with teammates

Cost Savings

  • No battery drain
  • No component wear
  • Test before building hardware
  • Identify design flaws early

​
Limitations of Simulation

Simulation is not reality! Always test on hardware:
  • Physics accuracy: Friction, slip, and dynamics differ from real world
  • Sensor noise: Simulated sensors are cleaner than reality
  • Latency: Real hardware has communication delays
  • Power constraints: Simulation doesn’t model battery drain
  • Environmental factors: Lighting, temperature, floor texture not modeled
Golden rule: Simulate to develop, validate on hardware!

​
Next Steps

URDF/Xacro

Build robot model for simulation

Navigation Testing

Test Nav2 in simulation before hardware

SLAM

Create maps in simulated environment

Data Visualization

View simulation data in RViz

​
References