Environment Setup and Validation#

On this page we briefly describe the environment used in this example workflow and validate that we can load it in Isaac Lab.

Docker Container: Base (see Installation for more details)

./docker/run_docker.sh

Environment Description#

The static apple-to-plate workflow ships its own GalileoG1StaticPickAndPlaceEnvironment, exposed by the example-environment registry as galileo_g1_static_pick_and_place. It reuses the same galileo_locomanip shelf scene and the same OpenXR retargeter as the loco-manipulation apple-to-plate environment, but defaults to the g1_wbc_agile_pink embodiment for recording. The static task never walks, so AGILE’s single-policy lower-body backend is a better fit than HOMIE’s stand/walk model split. Both the apple and destination plate are placed on the same shelf, so the WBC only needs to hold the standing pose.

Note

Recording vs. evaluation embodiment. Teleoperation in Teleoperation Data Collection uses the default g1_wbc_agile_pink (PinkIK + AGILE), while closed-loop policy evaluation in Closed-Loop Policy Inference and Evaluation uses g1_wbc_agile_joint (direct joint control + AGILE). Both share the same AGILE lower-body backend; the _joint twin just bypasses PinkIK at inference because the policy is trained on the joint-space targets that PinkIK produced during recording. This matters for finetuning: see Policy Post-Training (GR00T N1.7) for the rationale.

Environment Composition#

The full implementation lives in isaaclab_arena_environments/galileo_g1_static_pick_and_place_environment.py. This page keeps the breakdown at the composition level so the docs stay focused on how an Arena environment is assembled instead of mirroring every placement constant or scene utility in the source file.

Component	Role in this workflow
Environment name	`galileo_g1_static_pick_and_place`
Background	`galileo_locomanip` provides the Galileo shelf scene reused from the loco-manipulation workflow.
Embodiment	`g1_wbc_agile_pink` is the recording default; `g1_wbc_agile_joint` is used for policy evaluation after training.
Pick-up object	`apple_01_objaverse_robolab` by default, selected through `--object`.
Destination	`clay_plates_hot3d_robolab` by default, selected through `--destination`.
Scene utilities	The registered environment handles tuned object placement, object scale, shelf support, finger contact friction, and background cleanup internally.
Task	`PickAndPlaceTask` checks whether the apple is moved onto the plate and provides the success metrics used later in evaluation.
Teleop device	`openxr` is attached only when `--teleop_device openxr` is passed, so the same environment can also be used for replay and policy evaluation without XR.

At a high level, defining an Arena environment is just selecting assets, composing a scene, choosing a task, and returning an IsaacLabArenaEnvironment:

background = self.asset_registry.get_asset_by_name("galileo_locomanip")()
pick_up_object = self.asset_registry.get_asset_by_name(args_cli.object)()
destination = self.asset_registry.get_asset_by_name(args_cli.destination)()
embodiment = self.asset_registry.get_asset_by_name(args_cli.embodiment)(
    enable_cameras=args_cli.enable_cameras
)
teleop_device = (
    self.device_registry.get_device_by_name(args_cli.teleop_device)()
    if args_cli.teleop_device is not None else None
)

scene = Scene(assets=[background, pick_up_object, destination])
task = PickAndPlaceTask(
    pick_up_object=pick_up_object,
    destination_location=destination,
    background_scene=background,
)

return IsaacLabArenaEnvironment(
    name=self.name,
    embodiment=embodiment,
    scene=scene,
    task=task,
    teleop_device=teleop_device,
)

The production environment adds task-specific defaults around this core composition: same-shelf poses for the apple and plate, a short episode timeout, an invisible shelf support surface, tuned contact/friction settings, and cleanup of a few reused background props. Those details are handled inside the registered environment so workflow users can run it through the CLI without editing the scene setup code.

See Concept Overview, Scene, and Task for more details on Arena environment composition.

Validate Environment with Automated Tests#

A dedicated pytest module covers the static apple-to-plate environment. It asserts that (i) the task does not terminate when the apple is at its initial pose, and (ii) the success termination fires after the apple is teleported above the plate and allowed to settle.

python -m pytest isaaclab_arena/tests/test_g1_static_pick_and_place.py -v

You should see both tests pass. They run headless with cameras enabled and take around a minute each. This is the fastest way to confirm the scene, task, and termination logic are wired up correctly without requiring teleoperation hardware.

Note

First-run asset cache: The Objaverse apple USD streams from an S3 staging bucket and PhysX re-cooks its collision mesh at the baked-in 0.01x scale. On a fresh machine the collision cook may land a frame after the first physics step, which can cause the apple to tunnel through the shelf surface on the very first load. Subsequent runs read the cached USD from /tmp/Assets/ and spawn correctly. If you see the apple fall through the shelf, just re-run the command once.