/mirvehicle_ws

Primary LanguageC++

mirvehicle_ws

This MIR(Mobile Intelligent Robotics Lab) Vehicle workspace has been prepared to create a working environment for an autonomous vehicle.

Contributors

  • Jaerock Kwon, Ph.D. Associate Professor
  • Shobit Sharma, Research Assistant
  • Mohamed Fasil Syed, Research Assistant

Folder Structure

  • mirvehicle: ROS nodes will be located.
  • mirvehicle_description: vehicle model. It includes urdf and meshes. There is a single launch file in launch folder. bulldogbolot_state_publisher.launch will start rviz with the config file, bulldogbolt_state_publisher.rviz. This is used to verify that the vehicle's urdf model are well made.
    • bulldogbolt: A specific vehicle model name. Under this folder, you must define a vehicle's model.
      • main.urdf.xacro: The main urdf. This will be called by launch files. This main urdf includes following three xacros.
      • properties.urdf.xacro: This defines the properties of the vehicle. You may need to change this to specify a vehicle model's properties.
      • structure.urdf.xacro: This defines the vehicle's body and wheels.
      • sensors.urdf.xacro: This defines cameras and laser sensors.
  • mirvheicle_gazebo: This is for Gazebo simulation.

    • config/bulldogbolt_control.yaml: controller defintions.

    • launch

      • bulldogbolt_empty.launch: Start the Gazebo server with Bulldogbolt model and a basic world. You need to start gzclient in a separate terminal if you want to use Gazebo.
      • bulldogbolt_gazebo.launch: Start the Gazebo server and client. This will start rviz as well to visualize sensor data. You can use a parameter input for a world name. Use world:=<world_name> where the world name can be found at world folder. The default world name is skidpan.
      • bulldogbolt_neighborhood.launch: This is same as bulldogbolot_gazebo.launch world:=neighborhood
      • mirvehicle.launch: This is a generic launch file that will be included by other launch files. There is no need to use this launch file directly.

    • models: 3D object models for Gazebo.

    • worlds: World files for Gazebo.

Acknowledgements

This project was possible with the generous works of Arizona University's the CAT Vehicle Testbed and Dataspeed Inc's ADAS system (dbw_mkz).