Skip to content

Latest commit

 

History

History
200 lines (131 loc) · 6.54 KB

README.md

File metadata and controls

200 lines (131 loc) · 6.54 KB
Raw

Andino Gazebo Simulation

📋 Description

This package provides a simulation environment for Andino in Gazebo Fortress relying on ros_gz to integrate it with ROS 2.

🗜️ Platforms

  • ROS 2: Humble Hawksbill
  • OS:
    • Ubuntu 22.04 Jammy Jellyfish
  • Gazebo:
    • Fortress

📥 Installation

From binaries

  1. Meet prerequisites: ROS 2 Humble installation
  2. Install andino_gz package 
    sudo apt install ros-humble-andino-gz
  3. You can jump directly to Usage section and start using it.

From source using a containerized environment

  1. Clone this repository
git clone [email protected]:ekumenlabs/andino_gz.git
  1. Set up docker environment: Refer to docker readme

Once the container is running and dependencies have been installed you can proceed to package building.

📦 Build

The package contains some dependencies that must be installed in order to build it:

rosdep install --from-paths src -i -y

Then build the package and source the install workspace. To do so run the following commands:

colcon build
source install/setup.bash

🚀 Usage

Andino simulation

Once the package is built and sourced, you can start a simulation.

ros2 launch andino_gz andino_gz.launch.py

Note: You can use world_name flag to indicate other world to use. (For example: depot.sdf(default), office.sdf, populated_office.sdf, empty.sdf)

By default the ros bridge and rviz are initialized. In case you prefer to disable any of those you can do it via its flags:

ros2 launch andino_gz andino_gz.launch.py ros_bridge:=False rviz:=False

To see a complete list of available arguments for the launch file do:

ros2 launch andino_gz andino_gz.launch.py --show-args

Using a different world:

ros2 launch andino_gz andino_gz.launch.py world_name:=populated_office.sdf

Make sure to review the required topics using ign topics and ros2 topic CLI tools. Also, consider using looking at the translation entries under andino_gz/config/bridge_config.yaml.

🤖🤖🤖 Multi robot simulation

This simulation also supports multi-robot simulation.

ros2 launch andino_gz andino_gz.launch.py robots:="
     andino1={x: 0.0, y: 0.0, z: 0.1, yaw: 0.};
     andino2={x: -0.4, y: 0.1, z: 0.1, yaw: 0.};
     andino3={x: -0.4, y: -0.1, z: 0.1, yaw: 0.};
     andino4={x: -0.8, y: 0.2, z: 0.1, yaw: 0.};
     andino5={x: -0.8, y: -0.2, z: 0.1, yaw: 0.};
     andino6={x: -0.8, y: 0.0, z: 0.1, yaw: 0.};"
  • Note1: You can add as many Andino as you want.
  • Note2: If enabling Nav2(Using nav2:=True) try first with 2 robots and see how they perform in your system before using more robots.

The launch file is in charge of:

  • Start Gazebo simulator with a defined world (See '--world_name' flag)
  • Spawn as many robots as commanded.
  • Launch ros bridge for each robot.
  • Launch Rviz visualization for each robot.

The simulation allows you to spawn as many robots as you want via the --robots flags. For that, you can pass the information of the robots in some YAML format via ROS2 cli:

  <robot_name>={x: 0.0, y: 0.0, yaw: 0.0, roll: 0.0, pitch: 0.0, yaw: 0.0};
  • Note1: A ROS Namespace is pushed for each robot so all the topics and nodes are called the same with a difference of a <robot_name> prefix.
  • Note2: For RViz2 to work with the ROS namespace that is pushed, one has to ensure that no absolute topics are being used.

🧭 Navigation

  1. Run the simulation passing the nav2 flag.

    ros2 launch andino_gz andino_gz.launch.py nav2:=True

    As before, you can launch as many robots as you want, for example launching two:

    ros2 launch andino_gz andino_gz.launch.py nav2:=True robots:="andino1={x: 0.0, y: 0.0, z: 0.1, yaw: 0.};andino2={x: 1.0, y: 0.0, z: 0.1, yaw: 0.};"

    Once Gazebo window pops up, play the simulation using the gui.

    Important!:

    • When using nav2 flag, the map must match the selected world. For example:
      • For depot.sdf world, the map need to be depot: 
        ros2 launch andino_gz andino_gz.launch.py nav2:=True world_name:=depot.sdf map:=depot
        Typically, as this is the default world-map, there is no need to indicate them.
      • For office.sdf and populated_office.sdf worlds, the map need to be office: 
        ros2 launch andino_gz andino_gz.launch.py nav2:=True world_name:=office.sdf map:=office

  2. An RViz window will be spawned for each robot so it can be controlled independently. Use 2D Pose Estimate to pass a hint to AMCL where is the initial point

    Note you have to do it per robot, namely, per RViz window.

  3. Send the desired goal using Nav2 Goal

🗺️ SLAM

  1. Run simulation with ros bridge and RViz.

    ros2 launch andino_gz andino_gz.launch.py

  2. Run slam toolbox

    ros2 launch andino_gz slam_toolbox_online_async.launch.py

    Configuration can be forwarded to the slam_toolbox_node. By default, the configuration parameters are obtained from andino's configuration file. In case a custom file is wanted to be passed, simply use the launch file argument for indicating the path to a new file.

    ros2 launch andino_gz slam_toolbox_online_async.launch.py slams_param_file:=<my_path>

  3. Visualize in RViz: Add map panel to RViz and see how the map is being generated.

🙌 Contributing

Issues or PRs are always welcome! Please refer to CONTRIBUTING doc.

Code development

Note that a Docker folder is provided for easy setting up the workspace.