### Verify UnitreeCppEnv Installation

Source: https://github.com/hansz8/robojudo/blob/release/docs/unitree_setup.md

Run this command to verify the installation of the UnitreeCppEnv. If an error occurs, check the installation of unitree_sdk2 and unitree_cpp.

```bash
python -c "from robojudo.environment import UnitreeCppEnv"
```

--------------------------------

### Verify UnitreeEnv Installation

Source: https://github.com/hansz8/robojudo/blob/release/docs/unitree_setup.md

Run this command to verify the installation of the UnitreeEnv. If an error occurs, check the unitree_sdk2py installation.

```bash
python -c "from robojudo.environment import UnitreeEnv"
```

--------------------------------

### Install RoboJuDo Modules

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Installs all required modules for RoboJuDo. Alternatively, specify individual modules to install using arguments.

```bash
# Install all required modules
python submodule_install.py

# Or specify modules to install with args
# python submodule_install.py unitree_cpp
```

--------------------------------

### Install UnitreeCpp Submodule

Source: https://github.com/hansz8/robojudo/blob/release/docs/unitree_setup.md

Install the unitree_cpp package as a submodule within RoboJuDo. Ensure unitree_sdk2 is installed first.

```bash
python submodule_install.py unitree_cpp
```

--------------------------------

### Install RoboJuDo

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Install RoboJuDo using pip. An optional step is included to install the CPU version of PyTorch for faster performance if a GPU is not available or preferred.

```bash
# Optional, install cpu version for speed up
pip install torch --index-url https://download.pytorch.org/whl/cpu
pip install -e .
```

--------------------------------

### Run RoboJuDo on Real Robot

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Starts the RoboJuDo pipeline on a real robot, using the 'g1_real' configuration. Assumes the robot is on the ground and prepared.

```bash
python scripts/run_pipeline.py -c g1_real
```

--------------------------------

### KeyboardCtrl Example Ctrl Data

Source: https://github.com/hansz8/robojudo/blob/release/docs/controller.md

Example structure for keyboard control data, including key presses and releases with timestamps. Used by KeyboardCtrl.

```text
[{'type': 'keyboard', 'name': 's', 'pressed': True, 'timestamp': 1758888074.643119}]
```

--------------------------------

### KeyboardCtrl Hotkey Configuration

Source: https://github.com/hansz8/robojudo/blob/release/docs/controller.md

Configure custom hotkeys for KeyboardCtrl to trigger specific commands. This example shows mapping keys and key combinations to commands.

```python
KeyboardCtrl(
  cfg_ctrl=KeyboardCtrlCfg(
      triggers_extra={
          "Key.space": "[TEST]",
          "\x01": "[CTRL_A]",
      }
    )
  )
```

--------------------------------

### JoystickCtrl Example Ctrl Data

Source: https://github.com/hansz8/robojudo/blob/release/docs/controller.md

Example structure for Xbox Joystick control data, including axes and button events. Used by JoystickCtrl.

```json
{"axes": {"LeftX": 0.0, "LeftY": 0.0, "RightX": 0.0, "RightY": 0.0, "LT": 0.0, "RT": 0.0}, "button_event": [{"type": "button", "name": "A", "pressed": False, "timestamp": 1758886189.6776087}]}
```

--------------------------------

### JoystickCtrl Hotkey Configuration

Source: https://github.com/hansz8/robojudo/blob/release/docs/controller.md

Configure custom hotkeys for JoystickCtrl to trigger specific commands. This example shows mapping button combinations to commands.

```python
JoystickCtrlCfg(
    triggers_extra={
        "RB+Down": "[POLICY_SWITCH],0",
        "LB+RB+A": "COMBO_TEST",
    }
),
```

--------------------------------

### Run ProtoMotions Tracker Pipeline (Simulation)

Source: https://github.com/hansz8/robojudo/blob/release/docs/policy.md

Launches the ProtoMotions tracker pipeline for MuJoCo simulation. Specify the motion path and index.

```bash
python scripts/run_tracker_pipeline.py -c g1_protomotions_tracker \
 --motion-path assets/motions/g1/g1_bones_seed_mini.pt \
 --motion-index 0
```

--------------------------------

### Run ProtoMotions Tracker Pipeline (Real Robot)

Source: https://github.com/hansz8/robojudo/blob/release/docs/policy.md

Launches the ProtoMotions tracker pipeline for real G1 robot deployment. Specify the motion path and index.

```bash
python scripts/run_tracker_pipeline.py -c g1_protomotions_tracker_real \
  --motion-path assets/motions/g1/g1_bones_seed_mini.pt \
  --motion-index 0
```

--------------------------------

### Run ProtoMotions Tracker Pipeline

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Execute the ProtoMotions tracker pipeline with specified motion path and index.

```bash
python scripts/run_tracker_pipeline.py -c g1_protomotions_tracker \
  --motion-path assets/motions/g1/g1_bones_seed_mini.pt \
  --motion-index 0

```

--------------------------------

### Run ProtoMotions Tracker Pipeline with Custom ONNX

Source: https://github.com/hansz8/robojudo/blob/release/docs/policy.md

Launches the ProtoMotions tracker pipeline, allowing you to specify a custom ONNX path for your exported tracker.

```bash
python scripts/run_tracker_pipeline.py -c g1_protomotions_tracker \
 --motion-path assets/motions/g1/g1_bones_seed_mini.pt \
 --motion-index 0 \
 --onnx-path /path/to/unified_pipeline.onnx
```

--------------------------------

### Deploy BeyondMimic and ASAP Policies

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Launches RoboJuDo with the BeyondMimic and ASAP policies. These provide advanced policy options for robot control.

```bash
python scripts/run_pipeline.py -c g1_beyondmimic
python scripts/run_pipeline.py -c g1_asap
```

--------------------------------

### Run RoboJuDo Simulation (Default)

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Launches the default G1 simulation environment for RoboJuDo. Requires an Xbox controller for navigation.

```bash
# run the default g1 sim2sim cfg
python scripts/run_pipeline.py
```

--------------------------------

### Configure UnitreeEnv for Real Robot Deployment (g1_real_env_cfg.py)

Source: https://github.com/hansz8/robojudo/blob/release/docs/unitree_setup.md

Update the global configuration for G1 in g1_real_env_cfg.py to specify the environment type and network interface for UnitreeEnv.

```python
class G1RealEnvCfg(G1EnvCfg, UnitreeEnvCfg):
    env_type: str = UnitreeEnvCfg.model_fields["env_type"].default
    # ====== ENV CONFIGURATION ======
    unitree: UnitreeEnvCfg.UnitreeCfg = UnitreeEnvCfg.UnitreeCfg(
        net_if="eth0", # EDIT HERE
        robot="g1",
        msg_type="hg",
    )
```

--------------------------------

### Run Loco-Mimic Pipeline Configurations

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Execute the Loco-Mimic pipeline with specified configurations for beyond-mimic and ASAP scenarios.

```bash
python scripts/run_pipeline.py -c g1_locomimic_beyondmimic

```

```bash
python scripts/run_pipeline.py -c g1_locomimic_asap

```

--------------------------------

### Configure UnitreeEnv for Real Robot Deployment (g1_cfg.py)

Source: https://github.com/hansz8/robojudo/blob/release/docs/unitree_setup.md

Modify the g1_real configuration in g1_cfg.py to set the environment type and network interface for UnitreeEnv.

```python
class g1_real(g1):
    env: G1RealEnvCfg = G1RealEnvCfg(
        env_type="UnitreeEnv",  # For unitree_sdk2py
        # env_type="UnitreeCppEnv", # For unitree_cpp, check README for more details
        unitree=UnitreeEnvCfg.UnitreeCfg(
            net_if="eth0",  # note: change to your network interface
            robot="g1",
            msg_type="hg",
        ),
    )
```

--------------------------------

### Run Multi-Policy Pipeline

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Executes the RoboJuDo pipeline with the 'g1_switch' configuration, enabling multi-policy switching between Unitree and AMO policies using an Xbox controller.

```bash
python scripts/run_pipeline.py -c g1_switch
```

--------------------------------

### Configure Real Robot Environment

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Python code snippet showing how to configure the environment for a real robot, specifically for the G1 model. It involves setting the environment type and network interface.

```python
class g1_real(g1):
    env: G1RealEnvCfg = G1RealEnvCfg(
        env_type="UnitreeEnv",  # For unitree_sdk2py
        # env_type="UnitreeCppEnv",  # For unitree_cpp, check README for more details
        unitree=G1UnitreeCfg(
            net_if="eth0",  # note: change to your network interface
        ),
    )
```

--------------------------------

### Clone Repository and Create Python Environment

Source: https://github.com/hansz8/robojudo/blob/release/README.md

Clone the RoboJuDo repository and set up a dedicated Conda environment for the project. This ensures dependency isolation.

```bash
git clone https://github.com/HansZ8/RoboJuDo.git
cd RoboJuDo/
# Example using conda
conda create -n robojudo python=3.11 -y
conda activate robojudo
```

--------------------------------

### Configure MotionCtrl with G1MotionCtrlCfg

Source: https://github.com/hansz8/robojudo/blob/release/docs/controller.md

Use G1MotionCtrlCfg to configure MotionCtrl. Specify the motion_name to load a specific motion file.

```python
G1MotionCtrlCfg(
    motion_name="amass_all",
)
```

--------------------------------

### Configure BeyondMimicCtrl with G1BeyondmimicCtrlCfg

Source: https://github.com/hansz8/robojudo/blob/release/docs/controller.md

Use G1BeyondmimicCtrlCfg to configure BeyondMimicCtrl. The motion_name parameter is used when the policy's use_motion_from_model is set to False.

```python
G1BeyondmimicCtrlCfg(
  motion_name="dance1_subject2", # only when policy: use_motion_from_model=False
)
```

--------------------------------

### UnitreePolicy Command Generation

Source: https://github.com/hansz8/robojudo/blob/release/docs/policy.md

Generates movement commands for the robot using joystick or Unitree controller input. Maps joystick axes to forward/backward, left/right, and turning commands.

```python
def _get_commands(self, ctrl_data: dict) -> list[float]:
    commands = np.zeros(3)
    for key in ctrl_data.keys():
        if key in ["JoystickCtrl", "UnitreeCtrl"]:
            axes = ctrl_data[key]["axes"]
            lx, ly, rx, ry = axes["LeftX"], axes["LeftY"], axes["RightX"], axes["RightY"]

            commands[0] = command_remap(ly, self.commands_map[0])
            commands[1] = command_remap(lx, self.commands_map[1])
            commands[2] = command_remap(rx, self.commands_map[2])
            break
    return commands
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.