Renames Unitree configs in locomotion tasks to match properly (#714)

# Description

There was a slight inconsistency-- we called `unitree_a1`,
`unitree_go1`, and at the same time `h1` and `g1`. This MR fixes this
inconsistency and adds some more docs.

## Type of change

- This change requires a documentation update

## Checklist

- [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with
`./isaaclab.sh --format`
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] I have added tests that prove my fix is effective or that my
feature works
- [x] I have updated the changelog and the corresponding version in the
extension's `config/extension.toml` file
- [x] I have added my name to the `CONTRIBUTORS.md` or my name already
exists there

source/extensions/omni.isaac.lab_tasks/config/extension.toml

 [package]
 
 # Note: Semantic Versioning is used: https://semver.org/
-version = "0.8.0"
+version = "0.8.1"
 
 # Description
 title = "Isaac Lab Environments"

source/extensions/omni.isaac.lab_tasks/docs/CHANGELOG.rst

 Changelog
 ---------
 
+0.8.1 (2024-08-02)
+~~~~~~~~~~~~~~~~~~
+
+Changed
+^^^^^^^
+
+* Renamed the folder names for Unitree robots in the manager-based locomotion tasks. Earlier, there was an inconsistency
+  in the folder names as some had ``unitree_`` prefix and some didn't. Now, none of the folders have the prefix.
+
+
 0.8.0 (2024-07-26)
 ~~~~~~~~~~~~~~~~~~
 

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/g1/rough_env_cfg.py

@@ -22,6 +22,8 @@ from omni.isaac.lab_assets import G1_MINIMAL_CFG  # isort: skip
 
 @configclass
 class G1Rewards(RewardsCfg):
+    """Reward terms for the MDP."""
+
     termination_penalty = RewTerm(func=mdp.is_terminated, weight=-200.0)
     track_lin_vel_xy_exp = RewTerm(
         func=mdp.track_lin_vel_xy_yaw_frame_exp,

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/h1/rough_env_cfg.py

@@ -22,6 +22,8 @@ from omni.isaac.lab_assets import H1_MINIMAL_CFG  # isort: skip
 
 @configclass
 class H1Rewards(RewardsCfg):
+    """Reward terms for the MDP."""
+
     termination_penalty = RewTerm(func=mdp.is_terminated, weight=-200.0)
     lin_vel_z_l2 = None
     track_lin_vel_xy_exp = RewTerm(

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/spot/README.md

-We would like to acknowledge The AI Institute's efforts in developing the Spot MDP from specifications provided by Boston Dynamics.
-They trained, verified, and deployed the resulting policy on the Spot hardware and demonstrated its capability and reliability out in the real world.
-The accompanying deployment code and access to Spot's low-level API will be available in the Spot RL Researcher Kit.
+# Acknowledgment
 
-We thank The AI Institute for their trailblazing use of and contributions to Isaac Lab and for sharing their code publicly with the community to promote wider use of the Nvidia RL ecosystem.
+We would like to acknowledge [The AI Institute](https://theaiinstitute.com/)'s efforts in developing
+the Spot RL environment from the specifications provided by Boston Dynamics.
+The team at The AI Institute trained, verified, and deployed the resulting policy on the Spot hardware.
+They demonstrated its capability and reliability out in the real world.
+
+The accompanying deployment code and access to Spot's low-level API is available with the [Spot RL
+Researcher Kit](https://bostondynamics.com/reinforcement-learning-researcher-kit/).

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/spot/flat_env_cfg.py

@@ -286,8 +286,8 @@ class SpotTerminationsCfg:
         func=mdp.illegal_contact,
         params={"sensor_cfg": SceneEntityCfg("contact_forces", body_names=["body", ".*leg"]), "threshold": 1.0},
     )
-    out_of_bounds = DoneTerm(
-        func=spot_mdp.terminations.terrain_out_of_bounds,
+    terrain_out_of_bounds = DoneTerm(
+        func=mdp.terrain_out_of_bounds,
         params={"asset_cfg": SceneEntityCfg("robot"), "distance_buffer": 3.0},
         time_out=True,
     )

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/spot/mdp/__init__.py

@@ -8,4 +8,3 @@
 
 from .events import *  # noqa: F401, F403
 from .rewards import *  # noqa: F401, F403
-from .terminations import *  # noqa: F401, F403

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/spot/mdp/rewards.py

@@ -3,6 +3,11 @@
 #
 # SPDX-License-Identifier: BSD-3-Clause
 
+"""This sub-module contains the reward functions that can be used for Spot's locomotion task.
+
+The functions can be passed to the :class:`omni.isaac.lab.managers.RewardTermCfg` object to
+specify the reward function and its parameters.
+"""
 
 from __future__ import annotations
 
@@ -18,7 +23,9 @@ if TYPE_CHECKING:
     from omni.isaac.lab.managers import RewardTermCfg
 
 
-# -- Task Rewards
+##
+# Task Rewards
+##
 
 
 def air_time_reward(
@@ -28,7 +35,7 @@ def air_time_reward(
     mode_time: float,
     velocity_threshold: float,
 ) -> torch.Tensor:
-    """Reward longer feet air and contact time"""
+    """Reward longer feet air and contact time."""
     # extract the used quantities (to enable type-hinting)
     contact_sensor: ContactSensor = env.scene.sensors[sensor_cfg.name]
     asset: Articulation = env.scene[asset_cfg.name]
@@ -146,6 +153,10 @@ class GaitReward(ManagerTermBase):
             torch.logical_or(cmd > 0.0, body_vel > self.velocity_threshold), sync_reward * async_reward, 0.0
         )
 
+    """
+    Helper functions.
+    """
+
     def _sync_reward_func(self, foot_0: int, foot_1: int) -> torch.Tensor:
         """Reward synchronization of two feet."""
         air_time = self.contact_sensor.data.current_air_time
@@ -177,7 +188,9 @@ def foot_clearance_reward(
     return torch.exp(-torch.sum(reward, dim=1) / std)
 
 
-# -- Regularization Penalties
+##
+# Regularization Penalties
+##
 
 
 def action_smoothness_penalty(env: ManagerBasedRLEnv) -> torch.Tensor:

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/mdp/__init__.py

@@ -9,3 +9,4 @@ from omni.isaac.lab.envs.mdp import *  # noqa: F401, F403
 
 from .curriculums import *  # noqa: F401, F403
 from .rewards import *  # noqa: F401, F403
+from .terminations import *  # noqa: F401, F403

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/mdp/rewards.py

@@ -3,6 +3,12 @@
 #
 # SPDX-License-Identifier: BSD-3-Clause
 
+"""Common functions that can be used to define rewards for the learning environment.
+
+The functions can be passed to the :class:`omni.isaac.lab.managers.RewardTermCfg` object to
+specify the reward function and its parameters.
+"""
+
 from __future__ import annotations
 
 import torch
@@ -61,6 +67,12 @@ def feet_air_time_positive_biped(env, command_name: str, threshold: float, senso
 
 
 def feet_slide(env, sensor_cfg: SceneEntityCfg, asset_cfg: SceneEntityCfg = SceneEntityCfg("robot")) -> torch.Tensor:
+    """Penalize feet sliding.
+
+    This function penalizes the agent for sliding its feet on the ground. The reward is computed as the
+    norm of the linear velocity of the feet multiplied by a binary contact sensor. This ensures that the
+    agent is penalized only when the feet are in contact with the ground.
+    """
     # Penalize feet sliding
     contact_sensor: ContactSensor = env.scene.sensors[sensor_cfg.name]
     contacts = contact_sensor.data.net_forces_w_history[:, :, sensor_cfg.body_ids, :].norm(dim=-1).max(dim=1)[0] > 1.0

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/config/spot/mdp/terminations.py → source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/locomotion/velocity/mdp/terminations.py

@@ -20,33 +20,33 @@ from omni.isaac.lab.managers import SceneEntityCfg
 if TYPE_CHECKING:
     from omni.isaac.lab.envs import ManagerBasedRLEnv
 
-"""
-Terrain size limits.
-"""
-
 
 def terrain_out_of_bounds(
     env: ManagerBasedRLEnv, asset_cfg: SceneEntityCfg = SceneEntityCfg("robot"), distance_buffer: float = 3.0
 ) -> torch.Tensor:
-    """Terminate when agents move too close to the edge of the terrain."""
-    # extract the used quantities (to enable type-hinting)
-    asset: RigidObject = env.scene[asset_cfg.name]
-
-    def get_map_size(env: ManagerBasedRLEnv) -> tuple[float, float]:
-        grid_width, grid_length = env.scene.terrain.cfg.terrain_generator.size
-        n_cols = env.scene.terrain.cfg.terrain_generator.num_cols
-        n_rows = env.scene.terrain.cfg.terrain_generator.num_rows
-        border_width = env.scene.terrain.cfg.terrain_generator.border_width
-        length = n_cols * grid_length + 2 * border_width
-        width = n_rows * grid_width + 2 * border_width
-        return (width, length)
+    """Terminate when the actor move too close to the edge of the terrain.
 
+    If the actor moves too close to the edge of the terrain, the termination is activated. The distance
+    to the edge of the terrain is calculated based on the size of the terrain and the distance buffer.
+    """
     if env.scene.cfg.terrain.terrain_type == "plane":
-        return False
+        return False  # we have infinite terrain because it is a plane
     elif env.scene.cfg.terrain.terrain_type == "generator":
-        map_width, map_height = get_map_size(env)
+        # obtain the size of the sub-terrains
+        terrain_gen_cfg = env.scene.terrain.cfg.terrain_generator
+        grid_width, grid_length = terrain_gen_cfg.size
+        n_rows, n_cols = terrain_gen_cfg.num_rows, terrain_gen_cfg.num_cols
+        border_width = terrain_gen_cfg.border_width
+        # compute the size of the map
+        map_width = n_rows * grid_width + 2 * border_width
+        map_height = n_cols * grid_length + 2 * border_width
+
+        # extract the used quantities (to enable type-hinting)
+        asset: RigidObject = env.scene[asset_cfg.name]
+
+        # check if the agent is out of bounds
         x_out_of_bounds = torch.abs(asset.data.root_pos_w[:, 0]) > 0.5 * map_width - distance_buffer
         y_out_of_bounds = torch.abs(asset.data.root_pos_w[:, 1]) > 0.5 * map_height - distance_buffer
         return torch.logical_or(x_out_of_bounds, y_out_of_bounds)
     else:
-        raise ValueError("Received unsupported terrain type, must be either 'plane' or 'generator'")
+        raise ValueError("Received unsupported terrain type, must be either 'plane' or 'generator'.")