README(EN).md

piper_sim

中文

ROS	STATE

1 Gazebo Simulation

1.1 Piper Gazebo Simulation (With Gripper)

Run the Gazebo simulation

cd piper_ros
source devel/setup.bash

roslaunch piper_gazebo piper_gazebo.launch

Control the gripper arm via RViz GUI (Run in a new terminal)

cd piper_ros
source devel/setup.bash

roslaunch piper_description display_gripper_urdf.launch

1.2 Piper Gazebo Simulation (Without Gripper)

Run the Gazebo simulation

cd piper_ros
source devel/setup.bash

roslaunch piper_gazebo piper_no_gripper_gazebo.launch

Control the arm without the gripper via RViz GUI (Run in a new terminal)

cd piper_ros
source devel/setup.bash

roslaunch piper_description display_no_gripper_urdf.launch

Control the arm in Piper Gazebo through the following topics

The /joint_states topic controls multiple joints simultaneously, while other topics control individual joints.

Principle: The joint_states_ctrl node converts /joint_states control information into control information for each joint, such as /piper_description/joint1_position_controller/command, to control the arm in Gazebo simulation. Note: Joint 7 directly controls both joint 7 and joint 8, so joint 8 is not involved in the control of /joint_states.

/joint_states
/gazebo/joint1_position_controller/command
/gazebo/joint2_position_controller/command
/gazebo/joint3_position_controller/command
/gazebo/joint4_position_controller/command
/gazebo/joint5_position_controller/command
/gazebo/joint6_position_controller/command
/gazebo/joint7_position_controller/command
/gazebo/joint8_position_controller/command

2 Mujoco Simulation

2.1 Installing Mujoco210 and mujoco-py

2.1.1 Install Mujoco

1. Download Mujoco210

2. Extract the files

   mkdir ~/.mujoco
   cd (Directory containing the package)
   tar -zxvf mujoco210-linux-x86_64.tar.gz -C ~/.mujoco

3. Add the environment variable

   echo "export LD_LIBRARY_PATH=~/.mujoco/mujoco210/bin:\$LD_LIBRARY_PATH" >> ~/.bashrc
   source ~/.bashrc

4. Test the installation

   cd ~/.mujoco/mujoco210/bin
   ./simulate ../model/humanoid.xml

2.1.2 Install mujoco-py

1. Download the source code

   git clone https://github.com/openai/mujoco-py.git

2. Install (This step can be done within a conda environment)

   cd ~/mujoco-py
   pip3 install -U 'mujoco-py<2.2,>=2.1'
   pip3 install -r requirements.txt
   pip3 install -r requirements.dev.txt
   python3 setup.py install
   sudo apt install libosmesa6-dev
   sudo apt install patchelf

3. Add the environment variable

   echo "export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/nvidia" >> ~/.bashrc
   source ~/.bashrc

4. Test the installation

   import mujoco_py
   import os
   mj_path = mujoco_py.utils.discover_mujoco()
   xml_path = os.path.join(mj_path, 'model', 'humanoid.xml')
   model = mujoco_py.load_model_from_path(xml_path)
   sim = mujoco_py.MjSim(model)
   print(sim.data.qpos)
   sim.step()
   print(sim.data.qpos)

2.2 Piper Mujoco Simulation (With Gripper)

Run the Mujoco simulation

cd piper_ros
source devel/setup.bash

roslaunch piper_mujoco piper_mujoco.launch

Control the gripper arm via RViz GUI (Run in a new terminal)

cd piper_ros
source devel/setup.bash

roslaunch piper_description display_gripper_urdf.launch

2.3 Piper Mujoco Simulation (Without Gripper)

Run the Mujoco simulation

cd piper_ros
source devel/setup.bash

roslaunch piper_mujoco piper_no_gripper_mujoco.launch

Control the arm without the gripper via RViz GUI (Run in a new terminal)

cd piper_ros
source devel/setup.bash

roslaunch piper_description display_no_gripper_urdf.launch

Control Parameter Explanation

Control parameters for the gripper

Control parameters for without gripper

• damping="100" Adjust joint damping

• kp="10000" Adjust joint control gain

• forcerange="-100 100" Adjust joint torque control (Set forcelimited="false" to true, then add forcerange="-100 100")

PID Control demo