This repository contains the necessary files to deploy policies on the Unitree Go1/Go2 + ARX5 mobile manipulation system.

• For Go1+ARX5 deployment, the process is similar to walk-these-way, using Docker for deployment with added ARX5 low-level interface code.
• For Go2+ARX5 deployment, it is more streamlined, with the low-level interfaces modified according to walk-these-ways-go2.

We’d like to express our gratitude for the open-source contributions that made this possible!

The system uses a remote control for speed control and a VR interface for controlling the end-effector pose. In the future, we plan to improve the system with joystick control for speed.

Contributions in the form of pull requests (PRs) are welcome!👏

We use the official controller of Unitree to control the speed of the quadruped robot.

We use Meta Quest 3 to control the end-effector pose of the ARX.

1. Streaming Meta Quest 3 with ALVR and SteamVR. We mainly use the left handle to control. The key mapping is as follows:

2. Move vr_streaming.py into your pc for streaming.

3. modify the ip + port in the vr_streaming.py and remote_pub.py.

   # NOTE This is the ip and port of the pc host connected to vr
   GLOBAL_IP = "192.168.12.198"
   GLOBAL_PORT = "34565"

1. Install the necessary dependencies

   sudo apt install can-utils
   sudo apt install net-tools
   sudo apt-get install libevdev-dev

2. Configure serial port mapping

   cd ~/go1_gym/go1_arx_deploy/unitree_legged_sdk_bin/arx
   # or 
   # >>> cd go2_arx_deploy/unitree_legged_sdk_bin/arx
   
   udevadm info -a -n /dev/ttyACM* | grep serial

   Example:

   
   
   

   Then copy the serial number and modify the arx_can.rules:

   # go1_arx_deploy/unitree_legged_sdk_bin/arx/arx_can.rules
   # or 
   # >>> go2_arx_deploy/unitree_legged_sdk_bin/arx_can.rules
   
   SUBSYSTEM=="tty", ATTRS{idVendor}=="16d0", ATTRS{idProduct}=="117e",
   ATTRS{serial}=="209738784D4D", SYMLINK+="canable0"

3. Setup

   sudo cp arx_can.rules /etc/udev/rules.d/
   sudo chmod 777 /etc/udev/rules.d/arx_can.rules
   
   sh reopen.sh
   sh reopen.sh  # better run twice

Start by following the instructions for Installing the Deployment Utility, which will guide you through downloading the Docker image. Then, transfer the code to the Go1 device with the IP address [email protected].

📌(Important) Follow these steps in the specified order for proper deployment .

Place the robot into damping mode. The control sequence is: [L2+A], [L2+B], [L1+L2+START]. After this, the robot should sit on the ground and the joints should move freely.

cd ~/go1_gym/go1_gym_deploy/unitree_legged_sdk_bin
sh make.sh
sudo ./lcm_position_vr

cd ~/go1_gym/go1_gym_deploy/docker
sudo make autostart

cd ~/go1_gym/go1_gym_deploy/
python3 remote_pub.py

The Go2 code can run on Nvidia Jetson Orin(optional) or PC. We run it on Nvidia Jetson Orin to achieve wireless operation.

Clone LCM repository to any location (where you usually place installed softwares), then install LCM:

git clone https://github.com/lcm-proj/lcm.git
mkdir build
cd build
cmake ..
make
sudo make install

Clone the lastest Unitree_SDK2 repository to any location, then install Unitree_SDK2:

git clone https://github.com/unitreerobotics/unitree_sdk2.git
cd unitree_sdk2

To build your own application with the SDK, you can install the unitree_sdk2 to your system directory:

mkdir build
cd build
cmake ..
sudo make install

(IMPORTANT) Follow these steps in the specified order for proper deployment:

cd go2_deployment/go2_gym/go2_gym_deploy/unitree_legged_sdk_bin
sh make.sh
sudo ./lcm_position_vr_go2 eth0

Replace eth0 with your own network interface address. According to the messages shown in terminal, press Enter for several times and the communication between LCM and unitree_sdk2 will set up.

cd go2_deployment
python3 go2_deployment/go2_gym/go2_gym_deploy/scripts/deploy_policy_vr.py

cd go2_deployment/go2_gym/go2_gym_deploy/
python3 remote_pub.py