Environment Setup and Validation

Docker Container: Base (see Installation for more details)

./docker/run_docker.sh

Environment Description

The dexsuite_lift Arena environment wraps the Isaac Lab Isaac-Lift-KukaAllegro MDP for evaluation. The physics backend defaults to Newton. Pass --presets physx to override that environment default.

The environment is defined in isaaclab_arena_environments/dexsuite_lift_environment.py:

The Dexsuite Lift Environment

class DexsuiteLiftEnvironment(ExampleEnvironmentBase):

    name: str = "dexsuite_lift"

    def get_env(self, args_cli: argparse.Namespace):
        import math

        import isaaclab_tasks.core.lift  # noqa: F401

        from isaaclab_arena.environments.isaaclab_arena_environment import IsaacLabArenaEnvironment
        from isaaclab_arena.scene.scene import Scene
        from isaaclab_arena.tasks.lift_object_task import DexsuiteLiftTask
        from isaaclab_arena.utils.physics_backend import PhysicsBackend
        from isaaclab_arena.utils.pose import Pose, PoseRange

        dexsuite_table = self.asset_registry.get_asset_by_name("procedural_table")()
        dexsuite_table.set_initial_pose(Pose(position_xyz=(-0.55, 0.0, 0.235)))

        manip_object = self.asset_registry.get_asset_by_name("procedural_cube")()
        manip_object.set_initial_pose(
            PoseRange(
                position_xyz_min=(-0.75, -0.1, 0.35),
                position_xyz_max=(-0.35, 0.3, 0.75),
                rpy_min=(-math.pi, -math.pi, -math.pi),
                rpy_max=(math.pi, math.pi, math.pi),
            )
        )

        ground_plane = self.asset_registry.get_asset_by_name("ground_plane")()
        light = self.asset_registry.get_asset_by_name("light")()

        embodiment = self.asset_registry.get_asset_by_name("kuka_allegro")()

        scene = Scene(assets=[dexsuite_table, manip_object, ground_plane, light])
        task = DexsuiteLiftTask(lift_object=manip_object, background_scene=dexsuite_table)

        dexsuite_rl_cfg_entry = (
            "isaaclab_tasks.core.lift.config.kuka_allegro.agents."
            "rsl_rl_ppo_cfg:KukaAllegroPPORunnerCfg"
        )

        return IsaacLabArenaEnvironment(
            name=self.name,
            embodiment=embodiment,
            scene=scene,
            task=task,
            teleop_device=None,
            rl_framework_entry_point="rsl_rl_cfg_entry_point",
            rl_policy_cfg=dexsuite_rl_cfg_entry,
            default_physics_backend=PhysicsBackend.NEWTON,
        )

Note

The environment declares Newton as its default backend. The common --presets CLI flag can override that default.

Step-by-Step Breakdown

1. Embodiment: Kuka Allegro

embodiment = self.asset_registry.get_asset_by_name("kuka_allegro")()

The KukaAllegroEmbodiment provides:

• Scene: Kuka LBR iiwa arm + Allegro Hand articulation, plus four fingertip contact sensors (index_link_3, middle_link_3, ring_link_3, thumb_link_3).

• Actions: Relative joint position control for all 23 joints (scale=0.1).

• Observations (three groups, each with history_length=5):

  • policy: object quaternion, target pose command, last action.

  • proprio: joint positions, joint velocities, hand-tip body states (palm + fingertips), fingertip contact forces.

  • perception: object point cloud (64 points, flattened).

• Events: Joint-position randomization on reset (±0.5 rad offset from default).

2. Scene and Task

scene = Scene(assets=[dexsuite_table, manip_object, ground_plane, light])
task = DexsuiteLiftTask(lift_object=manip_object, background_scene=dexsuite_table)

DexsuiteLiftTask is evaluation-only: no rewards, no curriculum. It provides the object_pose command (position-only, resampled every 2–3 s), a success termination at 5 cm position tolerance, and a time-out termination.

3. Physics Backend Selection

The physics backend is selected by ArenaEnvBuilder:

• Default (Newton): no extra flag needed.

• PhysX override: pass --presets physx to policy_runner.py.

When Newton is resolved, the builder automatically:

1. Applies the ArenaPhysicsCfg().newton configuration (MuJoCo-Warp solver with tuned parameters for dexterous manipulation).

2. Enables scene.replicate_physics = True (required by Newton).

Validation: Run Zero-Action Policy

Verify the environment loads correctly with a zero-action policy:

# PhysX override:
python isaaclab_arena/evaluation/policy_runner.py \
  --viz kit \
  --presets physx \
  --policy_type zero_action \
  --num_steps 100 \
  dexsuite_lift

# Newton (environment default):
python isaaclab_arena/evaluation/policy_runner.py \
  --viz newton \
  --policy_type zero_action \
  --num_steps 100 \
  dexsuite_lift

You should see the Kuka Allegro hand in the scene with the cuboid on the table.

Tip

--viz newton uses the MuJoCo viewer; --viz kit uses the Kit viewer. The visualizer setting is independent of the physics backend. For example, --viz kit --presets newton runs Newton physics with the Kit viewer.