We have developed a robotic simulation environment for lunar sampling based on the ROS platform. The simulation environment includes models of the lunar surface terrain, sampling objects, a lunar sampling car, and a robotic arm. These models are integrated into Gazebo for 3D dynamic simulation. Subsequently, we simulate the automatic sampling process of the rover. The final step involves controlling the robot car to move to a designated point and place the sampled objects. The entire sampling experiment is conducted through joint simulation using both rviz and Gazebo. Additionally, we use Blender to create the simulated lunar surface model. Our work is based on the respository https://github.com/wdkhx/ros_car_with_arm.

Our testing environment: Ubuntu 20.04

mkdir -p cakin_ws/src
cd catkin_ws/src/
git clone https://github.com/Iris1026/A-Remote-Controlled-Robot-For-Lunar-Sampling.git
cd ..
catkin_make
echo "source /home/XXX/catkin_ws/devel/set up.bash">> ˜/.bashrc
source ~/.bashrc
code .

Our lunar surface Model:

We have placed the lunar surface model created with Blender in the folder named "surface_model." If you wish to integrate our pre-made lunar surface terrain into Gazebo, you can refer to the link provided. Enter link here

If you prefer to customize the terrain, you can download Blender from its official website. Enter link here

Run launch file to start：

roslaunch marm_gazebo arm_bringup_moveit.launch

Insert the lunar surface model into Gazebo.

To clearly observe the lunar surface model, set the ground_plane to wireframe:

Run the automatic grabbing script：

rosrun function moveit_fk_demo.py

Execute the keyboard control script:

rosrun teleop_twist_keyboard teleop_twist_keyboard.py

Automatic grabbing effect：

Controlling the rover to navigate around obstacles：

Successfully reaching the target location：

Controlling the robotic arm joints via rviz：

Successfully placing the sampled objects：