Teleop Project (x86_64 PC)
Use venv/teleop for Pico / XR body tracking, realtime retarget inspection, and motion recording on a laptop or desktop.
Setup
uv --project venv/teleop sync
Install XRoboToolkit PC Service
Download the .deb from:
https://github.com/XR-Robotics/XRoboToolkit-PC-Service/releases
sudo apt install -y ./XRoboToolkit_PC_Service_*.deb
Start XRoboToolkit / XRobot from the desktop or app launcher after installation.
Clone the Pico SDK repos
mkdir -p external
git clone https://github.com/YanjieZe/XRoboToolkit-PC-Service-Pybind.git \
external/XRoboToolkit-PC-Service-Pybind
git clone https://github.com/XR-Robotics/XRoboToolkit-PC-Service.git \
external/XRoboToolkit-PC-Service
Build and install xrobotoolkit_sdk
bash scripts/setup/setup_xrobot_pybind.sh
The script expects the two repos above to exist under external/.
Enable Pico streaming
- Wear the leg trackers.
- Finish whole-body tracking calibration.
- Open
XRoboToolkit, connect to this machine, and enable whole-body streaming.
Verify Installation
Start the live retarget publisher:
uv --project venv/teleop run sim2real/teleop/pico_retarget_pub.py \
--bind tcp://*:28701 \
--publish_hz 50 \
--actual_human_height 1.80
In a second terminal, run the realtime viewer:
uv --project venv/teleop run sim2real/teleop/realtime_viewer.py \
--connect tcp://127.0.0.1:28701 \
--viewer_hz 50
If the viewer updates with live G1 retargeted motion, the teleop stack is ready.