[Newton] Add Warp based inhand_manipulation env

CONTRIBUTORS.md

@@ -74,6 +74,7 @@ Guidelines for modifications:
 * Jeonghwan Kim
 * Jia Lin Yuan
 * Jiakai Zhang
+* Jichuan Hu
 * Jinghuan Shang
 * Jingzhou Liu
 * Jinqi Wei

scripts/reinforcement_learning/rsl_rl/train.py

@@ -32,6 +32,12 @@
     "--distributed", action="store_true", default=False, help="Run training with multiple GPUs or nodes."
 )
 parser.add_argument("--export_io_descriptors", action="store_true", default=False, help="Export IO descriptors.")
+parser.add_argument(
+    "--timer",
+    action=argparse.BooleanOptionalAction,
+    default=False,
+    help="Enable Isaac Lab timers (use --no-timer to disable).",
+)
 # append RSL-RL cli arguments
 cli_args.add_rsl_rl_args(parser)
 # append AppLauncher cli args
@@ -83,8 +89,8 @@
 
 from isaaclab.utils.timer import Timer
 
-Timer.enable = False
-Timer.enable_display_output = False
+Timer.enable = args_cli.timer
+Timer.enable_display_output = args_cli.timer
 
 import isaaclab_tasks_experimental  # noqa: F401
 

source/isaaclab_tasks_experimental/isaaclab_tasks_experimental/direct/allegro_hand/__init__.py

@@ -0,0 +1,30 @@
+# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+
+"""
+Allegro Inhand Manipulation environment.
+"""
+
+import gymnasium as gym
+
+from . import agents
+
+##
+# Register Gym environments.
+##
+
+inhand_task_entry = "isaaclab_tasks_experimental.direct.inhand_manipulation"
+
+gym.register(
+    id="Isaac-Repose-Cube-Allegro-Direct-Warp-v0",
+    entry_point=f"{inhand_task_entry}.inhand_manipulation_warp_env:InHandManipulationWarpEnv",
+    disable_env_checker=True,
+    kwargs={
+        "env_cfg_entry_point": f"{__name__}.allegro_hand_warp_env_cfg:AllegroHandWarpEnvCfg",
+        "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
+        "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:AllegroHandPPORunnerCfg",
+        "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
+    },
+)

source/isaaclab_tasks_experimental/isaaclab_tasks_experimental/direct/allegro_hand/agents/__init__.py

@@ -0,0 +1,4 @@
+# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause

source/isaaclab_tasks_experimental/isaaclab_tasks_experimental/direct/allegro_hand/agents/rl_games_ppo_cfg.yaml

@@ -0,0 +1,91 @@
+# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+
+params:
+  seed: 42
+
+  # environment wrapper clipping
+  env:
+    # added to the wrapper
+    clip_observations: 5.0
+    # can make custom wrapper?
+    clip_actions: 1.0
+
+  algo:
+    name: a2c_continuous
+
+  model:
+    name: continuous_a2c_logstd
+
+  # doesn't have this fine grained control but made it close
+  network:
+    name: actor_critic
+    separate: False
+    space:
+      continuous:
+        mu_activation: None
+        sigma_activation: None
+
+        mu_init:
+          name: default
+        sigma_init:
+          name: const_initializer
+          val: 0
+        fixed_sigma: True
+    mlp:
+      units: [1024, 512, 256, 128]
+      activation: elu
+      d2rl: False
+
+      initializer:
+        name: default
+      regularizer:
+        name: None
+
+  load_checkpoint: False # flag which sets whether to load the checkpoint
+  load_path: '' # path to the checkpoint to load
+
+  config:
+    name: allegro_hand
+    env_name: rlgpu
+    device: 'cuda:0'
+    device_name: 'cuda:0'
+    multi_gpu: False
+    ppo: True
+    mixed_precision: False
+    normalize_input: True
+    normalize_value: True
+    value_bootstrap: True
+    num_actors: -1  # configured from the script (based on num_envs)
+    reward_shaper:
+      scale_value: 0.01
+    normalize_advantage: True
+    gamma: 0.99
+    tau : 0.95
+    learning_rate: 5e-4
+    lr_schedule: adaptive
+    schedule_type: standard
+    kl_threshold: 0.016
+    score_to_win: 100000
+    max_epochs: 5000
+    save_best_after: 100
+    save_frequency: 200
+    print_stats: True
+    grad_norm: 1.0
+    entropy_coef: 0.0
+    truncate_grads: True
+    e_clip: 0.2
+    horizon_length: 16
+    minibatch_size: 32768
+    mini_epochs: 5
+    critic_coef: 4
+    clip_value: True
+    seq_length: 4
+    bounds_loss_coef: 0.0001
+
+    player:
+      deterministic: True
+      games_num: 100000
+      print_stats: True

source/isaaclab_tasks_experimental/isaaclab_tasks_experimental/direct/allegro_hand/agents/rsl_rl_ppo_cfg.py

@@ -0,0 +1,38 @@
+# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+
+from isaaclab.utils import configclass
+
+from isaaclab_rl.rsl_rl import RslRlOnPolicyRunnerCfg, RslRlPpoActorCriticCfg, RslRlPpoAlgorithmCfg
+
+
+@configclass
+class AllegroHandPPORunnerCfg(RslRlOnPolicyRunnerCfg):
+    num_steps_per_env = 16
+    max_iterations = 5000
+    save_interval = 250
+    experiment_name = "allegro_hand"
+    policy = RslRlPpoActorCriticCfg(
+        init_noise_std=1.0,
+        actor_obs_normalization=True,
+        critic_obs_normalization=True,
+        actor_hidden_dims=[1024, 512, 256, 128],
+        critic_hidden_dims=[1024, 512, 256, 128],
+        activation="elu",
+    )
+    algorithm = RslRlPpoAlgorithmCfg(
+        value_loss_coef=1.0,
+        use_clipped_value_loss=True,
+        clip_param=0.2,
+        entropy_coef=0.005,
+        num_learning_epochs=5,
+        num_mini_batches=4,
+        learning_rate=5.0e-4,
+        schedule="adaptive",
+        gamma=0.99,
+        lam=0.95,
+        desired_kl=0.016,
+        max_grad_norm=1.0,
+    )

source/isaaclab_tasks_experimental/isaaclab_tasks_experimental/direct/allegro_hand/agents/skrl_ppo_cfg.yaml

@@ -0,0 +1,85 @@
+# Copyright (c) 2022-2026, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
+# All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+
+seed: 42
+
+
+# Models are instantiated using skrl's model instantiator utility
+# https://skrl.readthedocs.io/en/latest/api/utils/model_instantiators.html
+models:
+  separate: False
+  policy:  # see gaussian_model parameters
+    class: GaussianMixin
+    clip_actions: False
+    clip_log_std: True
+    min_log_std: -20.0
+    max_log_std: 2.0
+    initial_log_std: 0.0
+    network:
+      - name: net
+        input: STATES
+        layers: [1024, 512, 256, 128]
+        activations: elu
+    output: ACTIONS
+  value:  # see deterministic_model parameters
+    class: DeterministicMixin
+    clip_actions: False
+    network:
+      - name: net
+        input: STATES
+        layers: [1024, 512, 256, 128]
+        activations: elu
+    output: ONE
+
+
+# Rollout memory
+# https://skrl.readthedocs.io/en/latest/api/memories/random.html
+memory:
+  class: RandomMemory
+  memory_size: -1  # automatically determined (same as agent:rollouts)
+
+
+# PPO agent configuration (field names are from PPO_DEFAULT_CONFIG)
+# https://skrl.readthedocs.io/en/latest/api/agents/ppo.html
+agent:
+  class: PPO
+  rollouts: 16
+  learning_epochs: 5
+  mini_batches: 4
+  discount_factor: 0.99
+  lambda: 0.95
+  learning_rate: 5.0e-04
+  learning_rate_scheduler: KLAdaptiveLR
+  learning_rate_scheduler_kwargs:
+    kl_threshold: 0.016
+  state_preprocessor: RunningStandardScaler
+  state_preprocessor_kwargs: null
+  value_preprocessor: RunningStandardScaler
+  value_preprocessor_kwargs: null
+  random_timesteps: 0
+  learning_starts: 0
+  grad_norm_clip: 1.0
+  ratio_clip: 0.2
+  value_clip: 0.2
+  clip_predicted_values: True
+  entropy_loss_scale: 0.0
+  value_loss_scale: 2.0
+  kl_threshold: 0.0
+  rewards_shaper_scale: 0.01
+  time_limit_bootstrap: False
+  # logging and checkpoint
+  experiment:
+    directory: "allegro_hand"
+    experiment_name: ""
+    write_interval: auto
+    checkpoint_interval: auto
+
+
+# Sequential trainer
+# https://skrl.readthedocs.io/en/latest/api/trainers/sequential.html
+trainer:
+  class: SequentialTrainer
+  timesteps: 80000
+  environment_info: log