# ddt_rl_isaacgym

**Repository Path**: ddtrobot/ddt_rl_isaacgym

## Basic Information

- **Project Name**: ddt_rl_isaacgym
- **Description**: Isaac Gym Environments for Legged Robots
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-06-16
- **Last Updated**: 2026-08-13

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README


### Installation ###
1. Create a new python virtual env with python 3.6, 3.7 or 3.8 (3.8 recommended)
2. Install pytorch 1.10 with cuda-11.3:
    - `pip3 install torch==1.10.0+cu113 torchvision==0.11.1+cu113 torchaudio==0.10.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html`
3. Install Isaac Gym
   - Download and install Isaac Gym Preview 3 (Preview 2 will not work!) from https://developer.nvidia.com/isaac-gym
   - `cd isaacgym/python && pip install -e .`
   - Try running an example `cd examples && python 1080_balls_of_solitude.py`
   - For troubleshooting check docs `isaacgym/docs/index.html`)
4. Clone robot description assets (into the repo root):
    ```bash
    git clone https://github.com/DDTRobot/ddt_ros2_control/
    ```
    After cloning, the repo should look like:
    ```
    ddt_rl_isaacgym/
    ├── algorithm/
    ├── configs/
    ├── ddt_ros2_control/   ← cloned here
    ├── modules/
    ├── resources/
    ├── runner/
    ├── scripts/
    └── utils/
    ```
5. Fix URDF mesh paths (replaces `package://` URIs with absolute paths so Isaac Gym can locate the mesh files):
    ```bash
    python scripts/fix_urdf_package_paths.py
    ```

### CODE STRUCTURE ###
1. Each environment is defined in an config file (`$ROBOT_$TERRAIN_config.py`) in `configs`, such as `d1_flat_config.py`. Task `d1_flat_config.py` as an example. The config file contains: 
    - `D1Flat` class: containing reward functions and env settings.
    - `D1FlatCfg` class: containing all the environment parameters for training.
    - `D1FlatCfg_Play` class: containing all the environment parameters for playing.
    - `D1FlatCfgPPO` class: containing all the training parameters.
2. Both env and config classes use inheritance.  
3. Each non-zero reward scale specified in `cfg` will add a function with a corresponding name to the list of elements which will be summed to get the total reward.  
4. Tasks must be registered using `task_registry.register(name, EnvClass, EnvConfig, TrainConfig)`. This is done in `$ROBOT/__init__.py`, but can also be done from outside of this repository.  

### Usage ###
1. Train:  
  ```python scripts/train.py --task=d1_flat```
    -  To run on CPU add following arguments: `--sim_device=cpu`, `--rl_device=cpu` (sim on CPU and rl on GPU is possible).
    -  To run headless (no rendering) add `--headless`.
    - **Important**: To improve performance, once the training starts press `v` to stop the rendering. You can then enable it later to check the progress.
    - The trained policy is saved in `./logs/<experiment_name>/<date_time>_<run_name>/model_<iteration>.pt`. Where `<experiment_name>` and `<run_name>` are defined in the train config.
    -  The following command line arguments override the values set in the config files:
     - --task TASK: Task name.
     - --resume:   Resume training from a checkpoint
     - --experiment_name EXPERIMENT_NAME: Name of the experiment to run or load.
     - --run_name RUN_NAME:  Name of the run.
     - --load_run LOAD_RUN:   Name of the run to load when resume=True. If -1: will load the last run.
     - --checkpoint CHECKPOINT:  Saved model checkpoint number. If -1: will load the last checkpoint.
     - --num_envs NUM_ENVS:  Number of environments to create.
     - --seed SEED:  Random seed.
     - --max_iterations MAX_ITERATIONS:  Maximum number of training iterations.
1. Play a trained policy:  
```python scripts/simple_play.py --task=d1_flat_play```
    - By default, the loaded policy is the last model of the last run of the experiment folder.
    - Other runs/model iteration can be selected by setting `load_run` and `checkpoint` in the train config.
    - Export policy file is saved in `./model.pt` and `./policy.onnx`

### Troubleshooting ###
1. If you get the following error: `ImportError: libpython3.8m.so.1.0: cannot open shared object file: No such file or directory`, do: `sudo apt install libpython3.8`. It is also possible that you need to do `export LD_LIBRARY_PATH=/path/to/libpython/directory` / `export LD_LIBRARY_PATH=/path/to/conda/envs/your_env/lib`(for conda user. Replace /path/to/ to the corresponding path.).

### Known Issues ###
1. The contact forces reported by `net_contact_force_tensor` are unreliable when simulating on GPU with a triangle mesh terrain. A workaround is to use force sensors, but the force are propagated through the sensors of consecutive bodies resulting in an undesirable behaviour. However, for a legged robot it is possible to add sensors to the feet/end effector only and get the expected results. When using the force sensors make sure to exclude gravity from the reported forces with `sensor_options.enable_forward_dynamics_forces`. Example:
```
    sensor_pose = gymapi.Transform()
    for name in feet_names:
        sensor_options = gymapi.ForceSensorProperties()
        sensor_options.enable_forward_dynamics_forces = False # for example gravity
        sensor_options.enable_constraint_solver_forces = True # for example contacts
        sensor_options.use_world_frame = True # report forces in world frame (easier to get vertical components)
        index = self.gym.find_asset_rigid_body_index(robot_asset, name)
        self.gym.create_asset_force_sensor(robot_asset, index, sensor_pose, sensor_options)
    (...)

    sensor_tensor = self.gym.acquire_force_sensor_tensor(self.sim)
    self.gym.refresh_force_sensor_tensor(self.sim)
    force_sensor_readings = gymtorch.wrap_tensor(sensor_tensor)
    self.sensor_forces = force_sensor_readings.view(self.num_envs, 4, 6)[..., :3]
    (...)

    self.gym.refresh_force_sensor_tensor(self.sim)
    contact = self.sensor_forces[:, :, 2] > 1.
```

## Acknowledgment

The project uses some code from the following open-source code repositories:
- [legged_gym](https://github.com/leggedrobotics/legged_gym)
- [LocomotionWithNP3O](https://github.com/zeonsunlightyu/LocomotionWithNP3O)

## Any Questions?

If you have any more questions, please create an issue in this repository.