Opt out of nested background physics reset for Galileo scene

[peterd-NV]

Summary

Nested background physics reset became enabled by default and opted the Galileo scene into a reset path that it does not use. The Galileo scene contains a dolly prim that identified as an articulation from its USD schema, but PhysX does not expose a corresponding tensor articulation. The user is informed of this via an error message that is not fatal.

This PR disables reset_nested_physics to restore prior behaviour and prevent users from getting this error message.

[greptile-apps]

Confidence Score: 5/5

The PR appears safe to merge with no actionable correctness, security, or repository-rule issues identified.

Summary

• Passes reset_nested_physics=False when constructing the galileo_locomanip background.
• Leaves object, destination, embodiment, and task construction unchanged.

_{Reviews (1) · Last reviewed commit: "disable nested phsyics reset on galileo ..."}

[arena-review-bot]

🤖 Isaac Lab-Arena Review Bot

Summary

One-line change passing reset_nested_physics=False when the Galileo locomanip background is built, to silence a non-fatal PhysX error caused by a dolly prim that carries ArticulationRootAPI in USD but has no tensor articulation. The kwarg does reach Background.__init__ correctly via LibraryBackground(**kwargs). My concerns are about scope and placement rather than mechanics.

Design, Boundaries & Scope

The framework already tolerates this exact failure: ResetBackgroundPhysics._initialize_asset in isaaclab_arena/terms/events.py matches the Failed to create articulation at: ... Please check PhysX logs. signature, skips that root, and logs a carb.log_warn. So the dolly is handled and what remains is log noise — but the opt-out is scene-wide, so it also removes reset coverage from every other nested dynamic body in galileo_locomanip.usd, which is the drift #1177 deliberately enabled this for. Worth confirming there is nothing else movable in that scene before trading resets for a quieter log.

Second, the flag is set at one call site. It describes the USD, not this environment, so isaaclab_arena/tests/test_locomanip_object_on_termination.py:29 — which builds galileo_locomanip directly and steps the sim — still hits the old path, and the test now exercises a different configuration than the shipped environment. Setting it on GalileoLocomanipBackground in isaaclab_arena/assets/background_library.py would make the fix apply wherever the asset is used.

Findings

🟡 Warning: isaaclab_arena_environments/galileo_g1_locomanip_pick_and_place_environment.py:92 — Whole-scene opt-out for what appears to be log noise; loses nested-body resets. (inline)
🟡 Warning: isaaclab_arena_environments/galileo_g1_locomanip_pick_and_place_environment.py:92 — Belongs on the GalileoLocomanipBackground asset, with a NOTE explaining the dolly. (inline)

Test Coverage

No test change. isaaclab_arena/tests/test_background_physics_reset.py covers the opt-out flag generically, so a new test for this line is not strictly required — but if the flag moves onto GalileoLocomanipBackground, a one-line assert that the asset reports reset_nested_physics is False would pin the intent cheaply, and test_locomanip_object_on_termination.py would then automatically stay in sync with the environment.

Verdict

Minor fixes needed