Go2Arm_sim2sim

KO | EN

gitlite — search

#javascript #python #ai #llm #golang #go #android #machine-learning #rust #awesome #kubernetes #awesome-list #security #mcp

(★ 105)

A simulation environment for verifying RL algorithms for legged manipulators (IsaacLab to Gazebo)

Go2Arm_sim2sim Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Go2Arm_sim2sim
[![ROS Noetic](https://img.shields.io/badge/ros-noetic-brightgreen.svg?logo=ros)](https://wiki.ros.org/noetic)
[![Ubuntu 20.04](https://img.shields.io/badge/Ubuntu-20.04-blue.svg?logo=ubuntu)](https://ubuntu.com/)
[![License](https://img.shields.io/badge/license-Apache2.0-yellow.svg?logo=apache)](https://opensource.org/license/apache-2-0)

🚀 Welcome to the **Go2Arm_sim2sim** platform! This project aims to provide an efficient and convenient simulation environment for verifying reinforcement learning algorithms for **legged manipulators**, assisting your research and development efforts.

> [!IMPORTANT]
> **🎉 A brand-new version is here!**
> 
> Go2Arm_sim2sim has been integrated into [**LeggedManip_Lab**](https://github.com/zzzJie-Robot/LeggedManip_Lab) — a unified framework covering the full pipeline from IsaacLab training → MuJoCo deployment → real-world transfer, with no extra Gazebo setup needed.
>
> 👉 **New users should use [LeggedManip_Lab](https://github.com/zzzJie-Robot/LeggedManip_Lab)**
> 
> 👉 This repository is archived — no further updates

---

[中文文档](https://github.com/zzzJie-Robot/Go2Arm_sim2sim/blob/main/README_CN.md "中文文档")

💡 sim2sim represents the migration from the IsaacLab platform to the Gazebo platform. The deployment on physical platforms (sim2real) will be introduced in future versions.

- **Base platform**: Unitree Go2 quadruped
- **Manipulator**: Interbotix WidowX 250s robotic arm

>**💻 System Requirements**  
>This project is built on `ros-noetic` and is recommended to be run on the `Ubuntu20.04` operating system.

>**📢 IsaacLab Simulation**  
>If you would like to conduct reinforcement learning training for legged robotic manipulators in IsaacLab, please visit:
>[Go2Arm_Lab](https://github.com/zzzJie-Robot/Go2Arm_Lab)

| IsaacLab Simulation | Gazebo Simulation |
|--------------------|-------------------|
| ![IsaacLab](https://github.com/zzzJie-Robot/Go2Arm_Lab/raw/main/video/IsaacLab.gif) | ![Gazebo](https://github.com/zzzJie-Robot/Go2Arm_Lab/raw/main/video/Gazebo.gif) |

For more videos, please visit my [Bilibili homepage](https://space.bilibili.com/400627082?spm_id_from=333.1007.0.0).

# Quick Start Guide 📦
## Step 1: Get the Code Repository
Clone the project code to your local machine by running the following command in the terminal:  
```
git clone https://github.com/zzzJie-Robot/Go2Arm_sim2sim.git
```

## Step 2: Install Dependencies
**ROS Dependency Installation**

Ensure that the following ROS dependencies are installed on your system. You can install them all at once with the following command:
```
sudo apt install ros-noetic-teleop-twist-keyboard ros-noetic-controller-interface ros-noetic-gazebo-ros-control ros-noetic-joint-state-controller ros-noetic-effort-controllers ros-noetic-joint-trajectory-controller ros-noetic-joy ros-noetic-ros-control ros-noetic-ros-controllers ros-noetic-controller-manager
```
**libtorch Library Deployment**

1.Choose a suitable directory (e.g., your workspace) and navigate to it:
```
cd <YOUR_DESIRED_PATH>
```
2.Download the libtorch library:：
```
wget https://download.pytorch.org/libtorch/cpu/libtorch-cxx11-abi-shared-with-deps-2.0.1%2Bcpu.zip
```
3.Unzip the file and set the environment variable:
```
unzip libtorch-cxx11-abi-shared-with-deps-2.0.1+cpu.zip -d ./
echo 'export Torch_DIR=<YOUR_DESIRED_PATH>/libtorch' >> ~/.bashrc
source ~/.bashrc
```

## Step 3: Compile the Project
Navigate to the project root directory and run the compilation command:

```
cd /path/to/Go2Arm_sim2sim
catkin_make
```

## Step 4: Run the Example Program
Open three terminals and execute the following commands in sequence:

**Terminal 1: Environment Initialization**
```
cd /path/to/Go2Arm_sim2sim
source ./devel/setup.bash
roslaunch go2arm_gazebo go2arm.launch
```
**Terminal 2: Load the Reinforcement Learning Model**
```
cd /path/to/Go2Arm_sim2sim
source ./devel/setup.bash
rosrun go2arm_gazebo main
```
**Terminal 3: Start Keyboard Control**
```
cd /path/to/Go2Arm_sim2sim
source ./devel/setup.bash
rosrun go2arm_gazebo keyboard_control
```

## Keyboard Control for Your Robot

| Keyboard | Robot Dog Movement       | Keyboard | Gripper Movement       | Keyboard | Gripper Pose                     | Keyboard | Gripper Action   |
| -------- | ------------------------ | -------- | ---------------------- | -------- | -------------------------------- | -------- | ---------------- |
| w        | Robot dog moves forward  | i        | Gripper moves forward  | 7        | Gripper rolls clockwise          | 1        | Gripper grips    |
| s        | Robot dog moves backward | k        | Gripper moves backward | 4        | Gripper rolls counterclockwise   | 2        | Gripper releases |
| a        | Robot dog moves left     | j        | Gripper moves left     | 8        | Gripper pitches clockwise        |          |                  |
| d        | Robot dog moves right    | l        | Gripper moves right    | 5        | Gripper pitches counterclockwise |          |                  |
| q        | Robot dog turns left     | u        | Gripper moves up       | 9        | Gripper yaws clockwise           |          |                  |
| e        | Robot dog turns right    | o        | Gripper moves down     | 6        | Gripper yaws counterclockwise    |          |                  |

## Add Your RL Model
To use your custom reinforcement learning model, simply rename your `.pt` model file to `policy.pt` and replace the existing file with the same name in the project. The path is as follows:
```
cd /path/to/Go2Arm_sim2sim/model
```

>Notice ⚠️  
>Please ensure that the order of the actions output by your model is consistent with the order in this project, and also match the PD parameters of the motors.

## Acknowledgements🙏
During the development of this project, we referred to the following excellent open-source projects. We would like to extend our sincere thanks to them:
- Joint controller part:：[rl_sar](https://github.com/fan-ziqi/rl_sar)
- Gripper grasp plugin:[gazebo_grasp_plugin](https://github.com/JenniferBuehler/gazebo-pkgs)

Go2Arm_sim2sim

Go2Arm_sim2sim Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Go2Arm_sim2sim

Go2Arm_sim2sim Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README